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

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

Neuron 545 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.7444. Min Act: -0.0001

Data Index: 1359570 (C4 (Web Text))

Max Activating Token Index: 656

Text #1

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6780. Min Act: -0.0000

Data Index: 282820 (C4 (Web Text))

Max Activating Token Index: 111

Text #2

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6803. Min Act: -0.0000

Data Index: 974727 (C4 (Web Text))

Max Activating Token Index: 497

Text #3

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6786. Min Act: -0.0001

Data Index: 1189288 (C4 (Web Text))

Max Activating Token Index: 320

Text #4

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6960. Min Act: -0.0001

Data Index: 36268 (C4 (Web Text))

Max Activating Token Index: 866

Text #5

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6904. Min Act: -0.0001

Data Index: 1885 (C4 (Web Text))

Max Activating Token Index: 885

Text #6

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6827. Min Act: -0.0000

Data Index: 891487 (C4 (Web Text))

Max Activating Token Index: 1015

Text #7

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.7226. Min Act: -0.0001

Data Index: 420404 (C4 (Web Text))

Max Activating Token Index: 293

Text #8

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6355. Min Act: -0.0000

Data Index: 1112872 (C4 (Web Text))

Max Activating Token Index: 582

Text #9

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6374. Min Act: -0.0000

Data Index: 321178 (C4 (Web Text))

Max Activating Token Index: 450

Text #10

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6738. Min Act: -0.0000

Data Index: 81544 (C4 (Web Text))

Max Activating Token Index: 401

Text #11

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6714. Min Act: -0.0000

Data Index: 325850 (C4 (Web Text))

Max Activating Token Index: 681

Text #12

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6344. Min Act: -0.0000

Data Index: 1048711 (C4 (Web Text))

Max Activating Token Index: 163

Text #13

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6293. Min Act: -0.0000

Data Index: 308100 (C4 (Web Text))

Max Activating Token Index: 851

Text #14

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6920. Min Act: -0.0001

Data Index: 1388616 (Python Code)

Max Activating Token Index: 853

Text #15

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6214. Min Act: -0.0000

Data Index: 803740 (C4 (Web Text))

Max Activating Token Index: 1016

Text #16

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.6594. Min Act: -0.0000

Data Index: 594754 (C4 (Web Text))

Max Activating Token Index: 112

Text #17

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.5934. Min Act: -0.0001

Data Index: 1442012 (Python Code)

Max Activating Token Index: 396

Text #18

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.5901. Min Act: -0.0000

Data Index: 54955 (C4 (Web Text))

Max Activating Token Index: 497

Text #19

Max Range: 0.7444. Min Range: -0.7444

Max Act: 0.5692. Min Act: -0.0000

Data Index: 1272811 (C4 (Web Text))

Max Activating Token Index: 676