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

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

Neuron 3188 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: 1.0377. Min Range: -1.0377

Max Act: 0.9927. Min Act: -0.0001

Data Index: 530651 (C4 (Web Text))

Max Activating Token Index: 14

Text #1

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.9786. Min Act: -0.0001

Data Index: 160545 (C4 (Web Text))

Max Activating Token Index: 79

Text #2

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.9684. Min Act: -0.0001

Data Index: 723913 (C4 (Web Text))

Max Activating Token Index: 22

Text #3

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.9659. Min Act: -0.0001

Data Index: 846276 (C4 (Web Text))

Max Activating Token Index: 569

Text #4

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8598. Min Act: -0.0001

Data Index: 100234 (C4 (Web Text))

Max Activating Token Index: 309

Text #5

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8593. Min Act: -0.0001

Data Index: 1304128 (C4 (Web Text))

Max Activating Token Index: 230

Text #6

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8817. Min Act: -0.0001

Data Index: 1144210 (C4 (Web Text))

Max Activating Token Index: 796

Text #7

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.9641. Min Act: -0.0001

Data Index: 1135376 (C4 (Web Text))

Max Activating Token Index: 156

Text #8

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8486. Min Act: -0.0001

Data Index: 1174141 (C4 (Web Text))

Max Activating Token Index: 715

Text #9

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8641. Min Act: -0.0001

Data Index: 242492 (C4 (Web Text))

Max Activating Token Index: 846

Text #10

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8065. Min Act: -0.0001

Data Index: 597981 (C4 (Web Text))

Max Activating Token Index: 975

Text #11

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8168. Min Act: -0.0001

Data Index: 1528509 (Python Code)

Max Activating Token Index: 42

Text #12

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8373. Min Act: -0.0001

Data Index: 1358282 (C4 (Web Text))

Max Activating Token Index: 26

Text #13

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8646. Min Act: -0.0001

Data Index: 518138 (C4 (Web Text))

Max Activating Token Index: 130

Text #14

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8322. Min Act: -0.0001

Data Index: 93874 (C4 (Web Text))

Max Activating Token Index: 272

Text #15

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8303. Min Act: -0.0001

Data Index: 744180 (C4 (Web Text))

Max Activating Token Index: 794

Text #16

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.7805. Min Act: -0.0001

Data Index: 1457308 (Python Code)

Max Activating Token Index: 90

Text #17

Max Range: 1.0377. Min Range: -1.0377

Max Act: 1.0377. Min Act: -0.0001

Data Index: 1520870 (Python Code)

Max Activating Token Index: 805

Text #18

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.8297. Min Act: -0.0001

Data Index: 887772 (C4 (Web Text))

Max Activating Token Index: 626

Text #19

Max Range: 1.0377. Min Range: -1.0377

Max Act: 0.7771. Min Act: -0.0001

Data Index: 447706 (C4 (Web Text))

Max Activating Token Index: 470