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

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

Neuron 3138 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.4113. Min Act: 0.0000

Data Index: 1599207 (Python Code)

Max Activating Token Index: 887

Text #1

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.3383. Min Act: -0.0000

Data Index: 1649911 (Python Code)

Max Activating Token Index: 146

Text #2

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.3344. Min Act: -0.0000

Data Index: 1406994 (Python Code)

Max Activating Token Index: 285

Text #3

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.3191. Min Act: 0.0000

Data Index: 408878 (C4 (Web Text))

Max Activating Token Index: 812

Text #4

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2856. Min Act: -0.0000

Data Index: 1261658 (C4 (Web Text))

Max Activating Token Index: 344

Text #5

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2594. Min Act: -0.0000

Data Index: 628568 (C4 (Web Text))

Max Activating Token Index: 785

Text #6

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2506. Min Act: -0.0000

Data Index: 1541067 (Python Code)

Max Activating Token Index: 365

Text #7

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2554. Min Act: -0.0000

Data Index: 404096 (C4 (Web Text))

Max Activating Token Index: 544

Text #8

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2430. Min Act: -0.0000

Data Index: 278826 (C4 (Web Text))

Max Activating Token Index: 459

Text #9

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2426. Min Act: 0.0000

Data Index: 125270 (C4 (Web Text))

Max Activating Token Index: 687

Text #10

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2378. Min Act: -0.0000

Data Index: 974057 (C4 (Web Text))

Max Activating Token Index: 796

Text #11

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2571. Min Act: -0.0000

Data Index: 1198601 (C4 (Web Text))

Max Activating Token Index: 417

Text #12

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2236. Min Act: -0.0000

Data Index: 107244 (C4 (Web Text))

Max Activating Token Index: 842

Text #13

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2364. Min Act: -0.0000

Data Index: 1620751 (Python Code)

Max Activating Token Index: 281

Text #14

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2296. Min Act: -0.0000

Data Index: 1055429 (C4 (Web Text))

Max Activating Token Index: 896

Text #15

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.1987. Min Act: 0.0000

Data Index: 384836 (C4 (Web Text))

Max Activating Token Index: 516

Text #16

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2101. Min Act: 0.0000

Data Index: 1376047 (Python Code)

Max Activating Token Index: 49

Text #17

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2102. Min Act: 0.0000

Data Index: 626433 (C4 (Web Text))

Max Activating Token Index: 858

Text #18

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2140. Min Act: -0.0000

Data Index: 975437 (C4 (Web Text))

Max Activating Token Index: 887

Text #19

Max Range: 0.4113. Min Range: -0.4113

Max Act: 0.2087. Min Act: -0.0000

Data Index: 534025 (C4 (Web Text))

Max Activating Token Index: 71