Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 504 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1412. Min Act: -0.0001

Data Index: 171147 (C4 (Web Text))

Max Activating Token Index: 661

Text #1

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1412. Min Act: -0.0001

Data Index: 1102456 (C4 (Web Text))

Max Activating Token Index: 789

Text #2

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1379. Min Act: -0.0001

Data Index: 466869 (C4 (Web Text))

Max Activating Token Index: 172

Text #3

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1350. Min Act: -0.0001

Data Index: 199242 (C4 (Web Text))

Max Activating Token Index: 725

Text #4

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1270. Min Act: -0.0001

Data Index: 518419 (C4 (Web Text))

Max Activating Token Index: 346

Text #5

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1182. Min Act: -0.0001

Data Index: 1103833 (C4 (Web Text))

Max Activating Token Index: 711

Text #6

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1174. Min Act: -0.0001

Data Index: 185129 (C4 (Web Text))

Max Activating Token Index: 762

Text #7

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1129. Min Act: -0.0001

Data Index: 742789 (C4 (Web Text))

Max Activating Token Index: 138

Text #8

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1076. Min Act: -0.0001

Data Index: 1328931 (C4 (Web Text))

Max Activating Token Index: 950

Text #9

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1001. Min Act: -0.0001

Data Index: 1058046 (C4 (Web Text))

Max Activating Token Index: 342

Text #10

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1097. Min Act: -0.0001

Data Index: 193986 (C4 (Web Text))

Max Activating Token Index: 942

Text #11

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1011. Min Act: -0.0001

Data Index: 12921 (C4 (Web Text))

Max Activating Token Index: 970

Text #12

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.1018. Min Act: -0.0001

Data Index: 331124 (C4 (Web Text))

Max Activating Token Index: 636

Text #13

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0942. Min Act: -0.0001

Data Index: 998224 (C4 (Web Text))

Max Activating Token Index: 490

Text #14

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0958. Min Act: -0.0001

Data Index: 38438 (C4 (Web Text))

Max Activating Token Index: 891

Text #15

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0897. Min Act: -0.0001

Data Index: 55335 (C4 (Web Text))

Max Activating Token Index: 764

Text #16

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0817. Min Act: -0.0001

Data Index: 517381 (C4 (Web Text))

Max Activating Token Index: 551

Text #17

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0843. Min Act: -0.0001

Data Index: 541436 (C4 (Web Text))

Max Activating Token Index: 850

Text #18

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0803. Min Act: -0.0000

Data Index: 1390045 (Python Code)

Max Activating Token Index: 260

Text #19

Max Range: 0.1412. Min Range: -0.1412

Max Act: 0.0864. Min Act: -0.0001

Data Index: 36163 (C4 (Web Text))

Max Activating Token Index: 642