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

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

Neuron 3499 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: 1.1393. Min Range: -1.1393

Max Act: 1.1393. Min Act: -0.0001

Data Index: 623759 (C4 (Web Text))

Max Activating Token Index: 973

Text #1

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.9329. Min Act: -0.0001

Data Index: 744819 (C4 (Web Text))

Max Activating Token Index: 614

Text #2

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.9580. Min Act: -0.0000

Data Index: 1295420 (C4 (Web Text))

Max Activating Token Index: 544

Text #3

Max Range: 1.1393. Min Range: -1.1393

Max Act: 1.0336. Min Act: -0.0000

Data Index: 1235441 (C4 (Web Text))

Max Activating Token Index: 285

Text #4

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.9403. Min Act: -0.0001

Data Index: 892724 (C4 (Web Text))

Max Activating Token Index: 496

Text #5

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.8543. Min Act: -0.0000

Data Index: 201557 (C4 (Web Text))

Max Activating Token Index: 381

Text #6

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.8942. Min Act: -0.0000

Data Index: 93754 (C4 (Web Text))

Max Activating Token Index: 203

Text #7

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.9387. Min Act: -0.0000

Data Index: 141602 (C4 (Web Text))

Max Activating Token Index: 252

Text #8

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.7888. Min Act: -0.0001

Data Index: 1507203 (Python Code)

Max Activating Token Index: 436

Text #9

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.8303. Min Act: -0.0001

Data Index: 285704 (C4 (Web Text))

Max Activating Token Index: 245

Text #10

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.7387. Min Act: -0.0001

Data Index: 875587 (C4 (Web Text))

Max Activating Token Index: 453

Text #11

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.7804. Min Act: -0.0000

Data Index: 125946 (C4 (Web Text))

Max Activating Token Index: 624

Text #12

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.7676. Min Act: -0.0001

Data Index: 615644 (C4 (Web Text))

Max Activating Token Index: 209

Text #13

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.7711. Min Act: -0.0000

Data Index: 467450 (C4 (Web Text))

Max Activating Token Index: 955

Text #14

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.6963. Min Act: -0.0001

Data Index: 217340 (C4 (Web Text))

Max Activating Token Index: 243

Text #15

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.6501. Min Act: -0.0001

Data Index: 1368703 (Python Code)

Max Activating Token Index: 228

Text #16

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.7357. Min Act: -0.0001

Data Index: 1428428 (Python Code)

Max Activating Token Index: 149

Text #17

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.6785. Min Act: -0.0000

Data Index: 1637475 (Python Code)

Max Activating Token Index: 254

Text #18

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.6769. Min Act: -0.0001

Data Index: 420988 (C4 (Web Text))

Max Activating Token Index: 900

Text #19

Max Range: 1.1393. Min Range: -1.1393

Max Act: 0.6879. Min Act: -0.0001

Data Index: 134751 (C4 (Web Text))

Max Activating Token Index: 355