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

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

Neuron 2365 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.0746. Min Range: -7.0746

Max Act: 7.0589. Min Act: -0.0001

Data Index: 1176280 (C4 (Web Text))

Max Activating Token Index: 624

Text #1

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.6088. Min Act: -0.0001

Data Index: 1000002 (C4 (Web Text))

Max Activating Token Index: 387

Text #2

Max Range: 7.0746. Min Range: -7.0746

Max Act: 7.0746. Min Act: -0.0000

Data Index: 655373 (C4 (Web Text))

Max Activating Token Index: 889

Text #3

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.8635. Min Act: -0.0001

Data Index: 536922 (C4 (Web Text))

Max Activating Token Index: 760

Text #4

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.6691. Min Act: -0.0000

Data Index: 1241658 (C4 (Web Text))

Max Activating Token Index: 819

Text #5

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.5320. Min Act: -0.0000

Data Index: 808625 (C4 (Web Text))

Max Activating Token Index: 372

Text #6

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.6083. Min Act: -0.0000

Data Index: 226115 (C4 (Web Text))

Max Activating Token Index: 192

Text #7

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.7157. Min Act: -0.0001

Data Index: 340199 (C4 (Web Text))

Max Activating Token Index: 282

Text #8

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.4614. Min Act: -0.0000

Data Index: 208583 (C4 (Web Text))

Max Activating Token Index: 440

Text #9

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.1144. Min Act: -0.0000

Data Index: 1251360 (C4 (Web Text))

Max Activating Token Index: 296

Text #10

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.9760. Min Act: -0.0001

Data Index: 1012107 (C4 (Web Text))

Max Activating Token Index: 973

Text #11

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.4856. Min Act: -0.0000

Data Index: 1008526 (C4 (Web Text))

Max Activating Token Index: 817

Text #12

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.6336. Min Act: -0.0001

Data Index: 1088838 (C4 (Web Text))

Max Activating Token Index: 735

Text #13

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.5752. Min Act: -0.0001

Data Index: 564315 (C4 (Web Text))

Max Activating Token Index: 711

Text #14

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.1962. Min Act: -0.0001

Data Index: 866483 (C4 (Web Text))

Max Activating Token Index: 569

Text #15

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.1888. Min Act: -0.0001

Data Index: 550130 (C4 (Web Text))

Max Activating Token Index: 30

Text #16

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.3485. Min Act: -0.0001

Data Index: 1207319 (C4 (Web Text))

Max Activating Token Index: 50

Text #17

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.1785. Min Act: -0.0001

Data Index: 1345050 (C4 (Web Text))

Max Activating Token Index: 761

Text #18

Max Range: 7.0746. Min Range: -7.0746

Max Act: 5.9488. Min Act: -0.0001

Data Index: 737614 (C4 (Web Text))

Max Activating Token Index: 51

Text #19

Max Range: 7.0746. Min Range: -7.0746

Max Act: 6.4565. Min Act: -0.0001

Data Index: 833446 (C4 (Web Text))

Max Activating Token Index: 675