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

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

Neuron 1598 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0320. Min Act: -0.0002

Data Index: 1055230 (C4 (Web Text))

Max Activating Token Index: 626

Text #1

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0283. Min Act: -0.0003

Data Index: 1614692 (Python Code)

Max Activating Token Index: 460

Text #2

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0268. Min Act: -0.0002

Data Index: 556780 (C4 (Web Text))

Max Activating Token Index: 717

Text #3

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0260. Min Act: -0.0003

Data Index: 1572149 (Python Code)

Max Activating Token Index: 78

Text #4

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0258. Min Act: -0.0003

Data Index: 448892 (C4 (Web Text))

Max Activating Token Index: 891

Text #5

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0251. Min Act: -0.0003

Data Index: 567314 (C4 (Web Text))

Max Activating Token Index: 34

Text #6

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0252. Min Act: -0.0003

Data Index: 417240 (C4 (Web Text))

Max Activating Token Index: 187

Text #7

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0252. Min Act: -0.0003

Data Index: 1134084 (C4 (Web Text))

Max Activating Token Index: 71

Text #8

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0249. Min Act: -0.0003

Data Index: 1342854 (C4 (Web Text))

Max Activating Token Index: 37

Text #9

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0240. Min Act: -0.0003

Data Index: 907264 (C4 (Web Text))

Max Activating Token Index: 235

Text #10

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0243. Min Act: -0.0003

Data Index: 1415346 (Python Code)

Max Activating Token Index: 380

Text #11

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0238. Min Act: -0.0003

Data Index: 698959 (C4 (Web Text))

Max Activating Token Index: 687

Text #12

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0243. Min Act: -0.0004

Data Index: 1019047 (C4 (Web Text))

Max Activating Token Index: 822

Text #13

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0230. Min Act: -0.0003

Data Index: 389966 (C4 (Web Text))

Max Activating Token Index: 364

Text #14

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0227. Min Act: -0.0002

Data Index: 544508 (C4 (Web Text))

Max Activating Token Index: 450

Text #15

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0225. Min Act: -0.0003

Data Index: 1571286 (Python Code)

Max Activating Token Index: 721

Text #16

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0229. Min Act: -0.0003

Data Index: 997097 (C4 (Web Text))

Max Activating Token Index: 727

Text #17

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0223. Min Act: -0.0003

Data Index: 1497307 (Python Code)

Max Activating Token Index: 610

Text #18

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0229. Min Act: -0.0003

Data Index: 1432082 (Python Code)

Max Activating Token Index: 815

Text #19

Max Range: 0.0320. Min Range: -0.0320

Max Act: 0.0211. Min Act: -0.0003

Data Index: 1629092 (Python Code)

Max Activating Token Index: 579