Model: SoLU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 2010 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-2l')

Text #0

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0323. Min Act: -0.0003

Data Index: 1020382 (C4 (Web Text))

Max Activating Token Index: 680

Text #1

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0296. Min Act: -0.0003

Data Index: 772760 (C4 (Web Text))

Max Activating Token Index: 572

Text #2

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0291. Min Act: -0.0003

Data Index: 855920 (C4 (Web Text))

Max Activating Token Index: 102

Text #3

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0278. Min Act: -0.0003

Data Index: 482194 (C4 (Web Text))

Max Activating Token Index: 573

Text #4

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0262. Min Act: -0.0004

Data Index: 519766 (C4 (Web Text))

Max Activating Token Index: 708

Text #5

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0270. Min Act: -0.0003

Data Index: 249060 (C4 (Web Text))

Max Activating Token Index: 682

Text #6

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0253. Min Act: -0.0003

Data Index: 1044306 (C4 (Web Text))

Max Activating Token Index: 467

Text #7

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0252. Min Act: -0.0003

Data Index: 24538 (C4 (Web Text))

Max Activating Token Index: 285

Text #8

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0240. Min Act: -0.0003

Data Index: 901757 (C4 (Web Text))

Max Activating Token Index: 729

Text #9

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0244. Min Act: -0.0003

Data Index: 653842 (C4 (Web Text))

Max Activating Token Index: 79

Text #10

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0243. Min Act: -0.0003

Data Index: 227607 (C4 (Web Text))

Max Activating Token Index: 730

Text #11

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0239. Min Act: -0.0003

Data Index: 47696 (C4 (Web Text))

Max Activating Token Index: 723

Text #12

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0242. Min Act: -0.0003

Data Index: 894167 (C4 (Web Text))

Max Activating Token Index: 421

Text #13

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0237. Min Act: -0.0003

Data Index: 1565903 (Python Code)

Max Activating Token Index: 313

Text #14

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0242. Min Act: -0.0003

Data Index: 1377300 (Python Code)

Max Activating Token Index: 201

Text #15

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0235. Min Act: -0.0003

Data Index: 1597807 (Python Code)

Max Activating Token Index: 256

Text #16

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0238. Min Act: -0.0003

Data Index: 974336 (C4 (Web Text))

Max Activating Token Index: 584

Text #17

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0236. Min Act: -0.0003

Data Index: 412038 (C4 (Web Text))

Max Activating Token Index: 158

Text #18

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0228. Min Act: -0.0003

Data Index: 589862 (C4 (Web Text))

Max Activating Token Index: 221

Text #19

Max Range: 0.0323. Min Range: -0.0323

Max Act: 0.0235. Min Act: -0.0004

Data Index: 1187616 (C4 (Web Text))

Max Activating Token Index: 535