Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 587 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6557. Min Act: -0.0002

Data Index: 466923 (C4 (Web Text))

Max Activating Token Index: 1008

Text #1

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7003. Min Act: -0.0002

Data Index: 707600 (C4 (Web Text))

Max Activating Token Index: 529

Text #2

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7398. Min Act: -0.0002

Data Index: 31738 (C4 (Web Text))

Max Activating Token Index: 207

Text #3

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7375. Min Act: -0.0002

Data Index: 69415 (C4 (Web Text))

Max Activating Token Index: 95

Text #4

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6955. Min Act: -0.0002

Data Index: 1289665 (C4 (Web Text))

Max Activating Token Index: 870

Text #5

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7054. Min Act: -0.0002

Data Index: 927520 (C4 (Web Text))

Max Activating Token Index: 336

Text #6

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7027. Min Act: -0.0002

Data Index: 455856 (C4 (Web Text))

Max Activating Token Index: 561

Text #7

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7160. Min Act: -0.0001

Data Index: 1123712 (C4 (Web Text))

Max Activating Token Index: 301

Text #8

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.7086. Min Act: -0.0002

Data Index: 1354168 (C4 (Web Text))

Max Activating Token Index: 220

Text #9

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6670. Min Act: -0.0002

Data Index: 108059 (C4 (Web Text))

Max Activating Token Index: 76

Text #10

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6582. Min Act: -0.0002

Data Index: 402872 (C4 (Web Text))

Max Activating Token Index: 967

Text #11

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6902. Min Act: -0.0002

Data Index: 1342884 (C4 (Web Text))

Max Activating Token Index: 634

Text #12

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6729. Min Act: -0.0002

Data Index: 1331621 (C4 (Web Text))

Max Activating Token Index: 828

Text #13

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6624. Min Act: -0.0002

Data Index: 18605 (C4 (Web Text))

Max Activating Token Index: 690

Text #14

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6676. Min Act: -0.0001

Data Index: 1121299 (C4 (Web Text))

Max Activating Token Index: 56

Text #15

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6496. Min Act: -0.0002

Data Index: 766935 (C4 (Web Text))

Max Activating Token Index: 187

Text #16

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6457. Min Act: -0.0002

Data Index: 126719 (C4 (Web Text))

Max Activating Token Index: 572

Text #17

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6322. Min Act: -0.0002

Data Index: 1320778 (C4 (Web Text))

Max Activating Token Index: 634

Text #18

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6416. Min Act: -0.0002

Data Index: 709003 (C4 (Web Text))

Max Activating Token Index: 284

Text #19

Max Range: 0.7398. Min Range: -0.7398

Max Act: 0.6099. Min Act: -0.0002

Data Index: 683207 (C4 (Web Text))

Max Activating Token Index: 807