Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2022 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.6324. Min Act: -0.0001

Data Index: 867801 (C4 (Web Text))

Max Activating Token Index: 24

Text #1

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5844. Min Act: -0.0001

Data Index: 563966 (C4 (Web Text))

Max Activating Token Index: 20

Text #2

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5547. Min Act: -0.0001

Data Index: 1071108 (C4 (Web Text))

Max Activating Token Index: 15

Text #3

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5455. Min Act: -0.0001

Data Index: 12733 (C4 (Web Text))

Max Activating Token Index: 33

Text #4

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5453. Min Act: -0.0001

Data Index: 520940 (C4 (Web Text))

Max Activating Token Index: 80

Text #5

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5400. Min Act: -0.0001

Data Index: 704429 (C4 (Web Text))

Max Activating Token Index: 48

Text #6

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5528. Min Act: -0.0001

Data Index: 572810 (C4 (Web Text))

Max Activating Token Index: 865

Text #7

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5520. Min Act: -0.0001

Data Index: 79497 (C4 (Web Text))

Max Activating Token Index: 108

Text #8

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5294. Min Act: -0.0002

Data Index: 65281 (C4 (Web Text))

Max Activating Token Index: 45

Text #9

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5480. Min Act: -0.0001

Data Index: 83165 (C4 (Web Text))

Max Activating Token Index: 183

Text #10

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5312. Min Act: -0.0001

Data Index: 1112591 (C4 (Web Text))

Max Activating Token Index: 44

Text #11

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5100. Min Act: -0.0001

Data Index: 617924 (C4 (Web Text))

Max Activating Token Index: 684

Text #12

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.4994. Min Act: -0.0001

Data Index: 74880 (C4 (Web Text))

Max Activating Token Index: 407

Text #13

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5012. Min Act: -0.0001

Data Index: 593626 (C4 (Web Text))

Max Activating Token Index: 167

Text #14

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5174. Min Act: -0.0001

Data Index: 937680 (C4 (Web Text))

Max Activating Token Index: 336

Text #15

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.4959. Min Act: -0.0001

Data Index: 902830 (C4 (Web Text))

Max Activating Token Index: 429

Text #16

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.5549. Min Act: -0.0001

Data Index: 1218297 (C4 (Web Text))

Max Activating Token Index: 891

Text #17

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.4877. Min Act: -0.0001

Data Index: 1224426 (C4 (Web Text))

Max Activating Token Index: 9

Text #18

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.4688. Min Act: -0.0001

Data Index: 1003752 (C4 (Web Text))

Max Activating Token Index: 693

Text #19

Max Range: 0.6324. Min Range: -0.6324

Max Act: 0.4817. Min Act: -0.0001

Data Index: 184035 (C4 (Web Text))

Max Activating Token Index: 426