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

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

Neuron 3644 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.0397. Min Range: -2.0397

Max Act: 2.0397. Min Act: -0.0002

Data Index: 1388406 (Python Code)

Max Activating Token Index: 960

Text #1

Max Range: 2.0397. Min Range: -2.0397

Max Act: 1.2501. Min Act: -0.0002

Data Index: 426015 (C4 (Web Text))

Max Activating Token Index: 940

Text #2

Max Range: 2.0397. Min Range: -2.0397

Max Act: 1.1805. Min Act: -0.0002

Data Index: 1520298 (Python Code)

Max Activating Token Index: 887

Text #3

Max Range: 2.0397. Min Range: -2.0397

Max Act: 1.0434. Min Act: -0.0002

Data Index: 627654 (C4 (Web Text))

Max Activating Token Index: 409

Text #4

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.8674. Min Act: -0.0002

Data Index: 912217 (C4 (Web Text))

Max Activating Token Index: 186

Text #5

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.9345. Min Act: -0.0002

Data Index: 1186189 (C4 (Web Text))

Max Activating Token Index: 472

Text #6

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.8135. Min Act: -0.0002

Data Index: 1344090 (C4 (Web Text))

Max Activating Token Index: 982

Text #7

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7079. Min Act: -0.0002

Data Index: 1561276 (Python Code)

Max Activating Token Index: 169

Text #8

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7501. Min Act: -0.0002

Data Index: 914857 (C4 (Web Text))

Max Activating Token Index: 634

Text #9

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7199. Min Act: -0.0002

Data Index: 1600897 (Python Code)

Max Activating Token Index: 244

Text #10

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7699. Min Act: -0.0002

Data Index: 1556244 (Python Code)

Max Activating Token Index: 662

Text #11

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.6706. Min Act: -0.0002

Data Index: 1577067 (Python Code)

Max Activating Token Index: 589

Text #12

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7399. Min Act: -0.0002

Data Index: 1475463 (Python Code)

Max Activating Token Index: 288

Text #13

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7073. Min Act: -0.0002

Data Index: 867415 (C4 (Web Text))

Max Activating Token Index: 174

Text #14

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7028. Min Act: -0.0002

Data Index: 651096 (C4 (Web Text))

Max Activating Token Index: 269

Text #15

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.6791. Min Act: -0.0002

Data Index: 1536805 (Python Code)

Max Activating Token Index: 342

Text #16

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.6910. Min Act: -0.0002

Data Index: 1656307 (Python Code)

Max Activating Token Index: 651

Text #17

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.7373. Min Act: -0.0002

Data Index: 443271 (C4 (Web Text))

Max Activating Token Index: 504

Text #18

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.6878. Min Act: -0.0002

Data Index: 1382786 (Python Code)

Max Activating Token Index: 764

Text #19

Max Range: 2.0397. Min Range: -2.0397

Max Act: 0.6521. Min Act: -0.0002

Data Index: 748807 (C4 (Web Text))

Max Activating Token Index: 160