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

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

Neuron 20 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: 0.0439. Min Range: -0.0439

Max Act: 0.0424. Min Act: -0.0003

Data Index: 788101 (C4 (Web Text))

Max Activating Token Index: 598

Text #1

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0439. Min Act: -0.0003

Data Index: 125576 (C4 (Web Text))

Max Activating Token Index: 333

Text #2

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0386. Min Act: -0.0003

Data Index: 663984 (C4 (Web Text))

Max Activating Token Index: 96

Text #3

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0359. Min Act: -0.0003

Data Index: 1100666 (C4 (Web Text))

Max Activating Token Index: 410

Text #4

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0349. Min Act: -0.0003

Data Index: 561271 (C4 (Web Text))

Max Activating Token Index: 205

Text #5

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0339. Min Act: -0.0003

Data Index: 773480 (C4 (Web Text))

Max Activating Token Index: 722

Text #6

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0329. Min Act: -0.0003

Data Index: 287251 (C4 (Web Text))

Max Activating Token Index: 989

Text #7

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0317. Min Act: -0.0003

Data Index: 873576 (C4 (Web Text))

Max Activating Token Index: 80

Text #8

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0312. Min Act: -0.0003

Data Index: 374490 (C4 (Web Text))

Max Activating Token Index: 864

Text #9

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0331. Min Act: -0.0003

Data Index: 323147 (C4 (Web Text))

Max Activating Token Index: 189

Text #10

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0334. Min Act: -0.0003

Data Index: 1061272 (C4 (Web Text))

Max Activating Token Index: 220

Text #11

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0323. Min Act: -0.0003

Data Index: 365395 (C4 (Web Text))

Max Activating Token Index: 54

Text #12

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0321. Min Act: -0.0003

Data Index: 70814 (C4 (Web Text))

Max Activating Token Index: 8

Text #13

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0323. Min Act: -0.0003

Data Index: 139588 (C4 (Web Text))

Max Activating Token Index: 305

Text #14

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0327. Min Act: -0.0003

Data Index: 935748 (C4 (Web Text))

Max Activating Token Index: 958

Text #15

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0324. Min Act: -0.0003

Data Index: 6883 (C4 (Web Text))

Max Activating Token Index: 366

Text #16

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0314. Min Act: -0.0003

Data Index: 12230 (C4 (Web Text))

Max Activating Token Index: 538

Text #17

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0305. Min Act: -0.0003

Data Index: 1246194 (C4 (Web Text))

Max Activating Token Index: 466

Text #18

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0305. Min Act: -0.0003

Data Index: 309304 (C4 (Web Text))

Max Activating Token Index: 137

Text #19

Max Range: 0.0439. Min Range: -0.0439

Max Act: 0.0313. Min Act: -0.0003

Data Index: 12714 (C4 (Web Text))

Max Activating Token Index: 48