Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4433 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.1277. Min Act: -0.0002

Data Index: 542780 (C4 (Web Text))

Max Activating Token Index: 455

Text #1

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.1842. Min Act: -0.0002

Data Index: 995986 (C4 (Web Text))

Max Activating Token Index: 1003

Text #2

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.0921. Min Act: -0.0002

Data Index: 714923 (C4 (Web Text))

Max Activating Token Index: 6

Text #3

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.1079. Min Act: -0.0002

Data Index: 321691 (C4 (Web Text))

Max Activating Token Index: 730

Text #4

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.1104. Min Act: -0.0002

Data Index: 1348858 (C4 (Web Text))

Max Activating Token Index: 958

Text #5

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.0528. Min Act: -0.0002

Data Index: 422876 (C4 (Web Text))

Max Activating Token Index: 22

Text #6

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.0888. Min Act: -0.0002

Data Index: 542828 (C4 (Web Text))

Max Activating Token Index: 126

Text #7

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.0838. Min Act: -0.0002

Data Index: 834785 (C4 (Web Text))

Max Activating Token Index: 69

Text #8

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9457. Min Act: -0.0002

Data Index: 1102677 (C4 (Web Text))

Max Activating Token Index: 187

Text #9

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.0214. Min Act: -0.0002

Data Index: 1174379 (C4 (Web Text))

Max Activating Token Index: 911

Text #10

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9702. Min Act: -0.0002

Data Index: 325841 (C4 (Web Text))

Max Activating Token Index: 355

Text #11

Max Range: 1.1842. Min Range: -1.1842

Max Act: 1.0029. Min Act: -0.0002

Data Index: 1116640 (C4 (Web Text))

Max Activating Token Index: 46

Text #12

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9682. Min Act: -0.0002

Data Index: 364143 (C4 (Web Text))

Max Activating Token Index: 614

Text #13

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9405. Min Act: -0.0002

Data Index: 1116748 (C4 (Web Text))

Max Activating Token Index: 105

Text #14

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9386. Min Act: -0.0002

Data Index: 894121 (C4 (Web Text))

Max Activating Token Index: 887

Text #15

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9848. Min Act: -0.0002

Data Index: 1150355 (C4 (Web Text))

Max Activating Token Index: 710

Text #16

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9549. Min Act: -0.0002

Data Index: 64512 (C4 (Web Text))

Max Activating Token Index: 725

Text #17

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9887. Min Act: -0.0002

Data Index: 832857 (C4 (Web Text))

Max Activating Token Index: 29

Text #18

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.9009. Min Act: -0.0002

Data Index: 1162540 (C4 (Web Text))

Max Activating Token Index: 533

Text #19

Max Range: 1.1842. Min Range: -1.1842

Max Act: 0.8443. Min Act: -0.0002

Data Index: 1276645 (C4 (Web Text))

Max Activating Token Index: 312