Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3207 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.5329. Min Act: -0.0000

Data Index: 772623 (C4 (Web Text))

Max Activating Token Index: 269

Text #1

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.5105. Min Act: -0.0000

Data Index: 205449 (C4 (Web Text))

Max Activating Token Index: 727

Text #2

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.5035. Min Act: -0.0000

Data Index: 1139008 (C4 (Web Text))

Max Activating Token Index: 384

Text #3

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4802. Min Act: -0.0000

Data Index: 1572819 (Python Code)

Max Activating Token Index: 924

Text #4

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4367. Min Act: -0.0000

Data Index: 1413853 (Python Code)

Max Activating Token Index: 782

Text #5

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4559. Min Act: -0.0000

Data Index: 1506698 (Python Code)

Max Activating Token Index: 624

Text #6

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4224. Min Act: -0.0000

Data Index: 172102 (C4 (Web Text))

Max Activating Token Index: 407

Text #7

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4346. Min Act: -0.0000

Data Index: 1372526 (Python Code)

Max Activating Token Index: 364

Text #8

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4302. Min Act: -0.0000

Data Index: 151970 (C4 (Web Text))

Max Activating Token Index: 384

Text #9

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4414. Min Act: -0.0000

Data Index: 430990 (C4 (Web Text))

Max Activating Token Index: 237

Text #10

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4194. Min Act: -0.0000

Data Index: 99578 (C4 (Web Text))

Max Activating Token Index: 1010

Text #11

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4187. Min Act: -0.0000

Data Index: 98973 (C4 (Web Text))

Max Activating Token Index: 307

Text #12

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4197. Min Act: -0.0000

Data Index: 1425910 (Python Code)

Max Activating Token Index: 835

Text #13

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4256. Min Act: -0.0000

Data Index: 1272803 (C4 (Web Text))

Max Activating Token Index: 634

Text #14

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4046. Min Act: -0.0000

Data Index: 1284167 (C4 (Web Text))

Max Activating Token Index: 461

Text #15

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.3515. Min Act: -0.0000

Data Index: 1553589 (Python Code)

Max Activating Token Index: 205

Text #16

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4204. Min Act: -0.0000

Data Index: 399643 (C4 (Web Text))

Max Activating Token Index: 858

Text #17

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.4011. Min Act: -0.0000

Data Index: 1494653 (Python Code)

Max Activating Token Index: 157

Text #18

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.3759. Min Act: -0.0000

Data Index: 661134 (C4 (Web Text))

Max Activating Token Index: 774

Text #19

Max Range: 0.5329. Min Range: -0.5329

Max Act: 0.3737. Min Act: -0.0000

Data Index: 612326 (C4 (Web Text))

Max Activating Token Index: 939