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

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

Neuron 2678 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0311. Min Range: -1.0311

Max Act: 1.0311. Min Act: -0.0000

Data Index: 1444789 (Python Code)

Max Activating Token Index: 173

Text #1

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.9788. Min Act: -0.0000

Data Index: 686046 (C4 (Web Text))

Max Activating Token Index: 469

Text #2

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8886. Min Act: -0.0000

Data Index: 1350231 (C4 (Web Text))

Max Activating Token Index: 638

Text #3

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8576. Min Act: -0.0000

Data Index: 1423852 (Python Code)

Max Activating Token Index: 65

Text #4

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8762. Min Act: -0.0000

Data Index: 119587 (C4 (Web Text))

Max Activating Token Index: 86

Text #5

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8082. Min Act: -0.0000

Data Index: 1168099 (C4 (Web Text))

Max Activating Token Index: 214

Text #6

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8777. Min Act: -0.0000

Data Index: 907146 (C4 (Web Text))

Max Activating Token Index: 498

Text #7

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8262. Min Act: -0.0000

Data Index: 1222219 (C4 (Web Text))

Max Activating Token Index: 329

Text #8

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8156. Min Act: -0.0000

Data Index: 1640984 (Python Code)

Max Activating Token Index: 746

Text #9

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8249. Min Act: -0.0000

Data Index: 83060 (C4 (Web Text))

Max Activating Token Index: 61

Text #10

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8239. Min Act: -0.0000

Data Index: 491010 (C4 (Web Text))

Max Activating Token Index: 481

Text #11

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7899. Min Act: -0.0000

Data Index: 1020974 (C4 (Web Text))

Max Activating Token Index: 930

Text #12

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7789. Min Act: -0.0000

Data Index: 859363 (C4 (Web Text))

Max Activating Token Index: 245

Text #13

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8035. Min Act: -0.0000

Data Index: 1346169 (C4 (Web Text))

Max Activating Token Index: 133

Text #14

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7743. Min Act: -0.0000

Data Index: 79247 (C4 (Web Text))

Max Activating Token Index: 444

Text #15

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7715. Min Act: -0.0000

Data Index: 973541 (C4 (Web Text))

Max Activating Token Index: 562

Text #16

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.8168. Min Act: -0.0000

Data Index: 1192701 (C4 (Web Text))

Max Activating Token Index: 355

Text #17

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7874. Min Act: -0.0000

Data Index: 1391877 (Python Code)

Max Activating Token Index: 583

Text #18

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7363. Min Act: -0.0000

Data Index: 1016198 (C4 (Web Text))

Max Activating Token Index: 183

Text #19

Max Range: 1.0311. Min Range: -1.0311

Max Act: 0.7787. Min Act: -0.0000

Data Index: 1231868 (C4 (Web Text))

Max Activating Token Index: 979