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

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

Neuron 3483 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.8288. Min Act: -0.0000

Data Index: 3007 (C4 (Web Text))

Max Activating Token Index: 432

Text #1

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.5513. Min Act: -0.0000

Data Index: 1178333 (C4 (Web Text))

Max Activating Token Index: 157

Text #2

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4944. Min Act: -0.0000

Data Index: 994558 (C4 (Web Text))

Max Activating Token Index: 586

Text #3

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.5160. Min Act: -0.0000

Data Index: 468024 (C4 (Web Text))

Max Activating Token Index: 205

Text #4

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4658. Min Act: -0.0000

Data Index: 800292 (C4 (Web Text))

Max Activating Token Index: 984

Text #5

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4691. Min Act: -0.0000

Data Index: 836425 (C4 (Web Text))

Max Activating Token Index: 642

Text #6

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4859. Min Act: -0.0000

Data Index: 50462 (C4 (Web Text))

Max Activating Token Index: 341

Text #7

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4840. Min Act: -0.0000

Data Index: 1289538 (C4 (Web Text))

Max Activating Token Index: 601

Text #8

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4694. Min Act: 0.0001

Data Index: 1134270 (C4 (Web Text))

Max Activating Token Index: 263

Text #9

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4124. Min Act: -0.0000

Data Index: 478007 (C4 (Web Text))

Max Activating Token Index: 647

Text #10

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4316. Min Act: -0.0000

Data Index: 47253 (C4 (Web Text))

Max Activating Token Index: 285

Text #11

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4406. Min Act: -0.0000

Data Index: 1200803 (C4 (Web Text))

Max Activating Token Index: 918

Text #12

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4301. Min Act: -0.0000

Data Index: 766438 (C4 (Web Text))

Max Activating Token Index: 789

Text #13

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4276. Min Act: -0.0000

Data Index: 1130531 (C4 (Web Text))

Max Activating Token Index: 838

Text #14

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4268. Min Act: 0.0000

Data Index: 664478 (C4 (Web Text))

Max Activating Token Index: 454

Text #15

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4097. Min Act: 0.0000

Data Index: 387896 (C4 (Web Text))

Max Activating Token Index: 38

Text #16

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4076. Min Act: -0.0000

Data Index: 241451 (C4 (Web Text))

Max Activating Token Index: 417

Text #17

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4343. Min Act: -0.0000

Data Index: 1268024 (C4 (Web Text))

Max Activating Token Index: 188

Text #18

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4108. Min Act: -0.0000

Data Index: 777950 (C4 (Web Text))

Max Activating Token Index: 121

Text #19

Max Range: 0.8288. Min Range: -0.8288

Max Act: 0.4088. Min Act: -0.0000

Data Index: 86769 (C4 (Web Text))

Max Activating Token Index: 593