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

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

Neuron 3712 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0365. Min Act: -0.0003

Data Index: 764860 (C4 (Web Text))

Max Activating Token Index: 236

Text #1

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0356. Min Act: -0.0003

Data Index: 339016 (C4 (Web Text))

Max Activating Token Index: 425

Text #2

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0326. Min Act: -0.0002

Data Index: 865519 (C4 (Web Text))

Max Activating Token Index: 983

Text #3

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0322. Min Act: -0.0002

Data Index: 1005217 (C4 (Web Text))

Max Activating Token Index: 115

Text #4

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0309. Min Act: -0.0002

Data Index: 64726 (C4 (Web Text))

Max Activating Token Index: 727

Text #5

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0276. Min Act: -0.0003

Data Index: 1647834 (Python Code)

Max Activating Token Index: 18

Text #6

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0280. Min Act: -0.0002

Data Index: 784370 (C4 (Web Text))

Max Activating Token Index: 38

Text #7

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0272. Min Act: -0.0003

Data Index: 221138 (C4 (Web Text))

Max Activating Token Index: 106

Text #8

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0256. Min Act: -0.0003

Data Index: 627200 (C4 (Web Text))

Max Activating Token Index: 393

Text #9

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0264. Min Act: -0.0002

Data Index: 1119895 (C4 (Web Text))

Max Activating Token Index: 265

Text #10

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0253. Min Act: -0.0003

Data Index: 1491134 (Python Code)

Max Activating Token Index: 931

Text #11

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0254. Min Act: -0.0002

Data Index: 586182 (C4 (Web Text))

Max Activating Token Index: 630

Text #12

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0264. Min Act: -0.0002

Data Index: 202949 (C4 (Web Text))

Max Activating Token Index: 881

Text #13

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0255. Min Act: -0.0002

Data Index: 312044 (C4 (Web Text))

Max Activating Token Index: 768

Text #14

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0233. Min Act: -0.0002

Data Index: 709088 (C4 (Web Text))

Max Activating Token Index: 773

Text #15

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0250. Min Act: -0.0003

Data Index: 252300 (C4 (Web Text))

Max Activating Token Index: 686

Text #16

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0246. Min Act: -0.0003

Data Index: 1144922 (C4 (Web Text))

Max Activating Token Index: 721

Text #17

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0246. Min Act: -0.0002

Data Index: 1083405 (C4 (Web Text))

Max Activating Token Index: 491

Text #18

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0251. Min Act: -0.0003

Data Index: 1540783 (Python Code)

Max Activating Token Index: 374

Text #19

Max Range: 0.0365. Min Range: -0.0365

Max Act: 0.0242. Min Act: -0.0002

Data Index: 625577 (C4 (Web Text))

Max Activating Token Index: 201