Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2018 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0300. Min Act: -0.0006

Data Index: 1280420 (C4 (Web Text))

Max Activating Token Index: 450

Text #1

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0265. Min Act: -0.0005

Data Index: 305185 (C4 (Web Text))

Max Activating Token Index: 871

Text #2

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0273. Min Act: -0.0006

Data Index: 17783 (C4 (Web Text))

Max Activating Token Index: 668

Text #3

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0263. Min Act: -0.0006

Data Index: 618182 (C4 (Web Text))

Max Activating Token Index: 85

Text #4

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0256. Min Act: -0.0006

Data Index: 184021 (C4 (Web Text))

Max Activating Token Index: 24

Text #5

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0261. Min Act: -0.0006

Data Index: 320546 (C4 (Web Text))

Max Activating Token Index: 689

Text #6

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0261. Min Act: -0.0005

Data Index: 803440 (C4 (Web Text))

Max Activating Token Index: 383

Text #7

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0254. Min Act: -0.0006

Data Index: 1348748 (C4 (Web Text))

Max Activating Token Index: 63

Text #8

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0255. Min Act: -0.0006

Data Index: 1147268 (C4 (Web Text))

Max Activating Token Index: 140

Text #9

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0253. Min Act: -0.0006

Data Index: 246108 (C4 (Web Text))

Max Activating Token Index: 706

Text #10

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0252. Min Act: -0.0006

Data Index: 982009 (C4 (Web Text))

Max Activating Token Index: 546

Text #11

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0255. Min Act: -0.0006

Data Index: 1193764 (C4 (Web Text))

Max Activating Token Index: 493

Text #12

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0254. Min Act: -0.0006

Data Index: 1058710 (C4 (Web Text))

Max Activating Token Index: 824

Text #13

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0242. Min Act: -0.0006

Data Index: 628522 (C4 (Web Text))

Max Activating Token Index: 170

Text #14

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0248. Min Act: -0.0006

Data Index: 1300536 (C4 (Web Text))

Max Activating Token Index: 227

Text #15

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0252. Min Act: -0.0006

Data Index: 108763 (C4 (Web Text))

Max Activating Token Index: 124

Text #16

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0262. Min Act: -0.0006

Data Index: 1207102 (C4 (Web Text))

Max Activating Token Index: 941

Text #17

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0245. Min Act: -0.0006

Data Index: 766997 (C4 (Web Text))

Max Activating Token Index: 891

Text #18

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0249. Min Act: -0.0006

Data Index: 1107620 (C4 (Web Text))

Max Activating Token Index: 734

Text #19

Max Range: 0.0300. Min Range: -0.0300

Max Act: 0.0238. Min Act: -0.0006

Data Index: 1039669 (C4 (Web Text))

Max Activating Token Index: 691