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

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

Neuron 3837 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.4512. Min Act: -0.0000

Data Index: 1616080 (Python Code)

Max Activating Token Index: 796

Text #1

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2770. Min Act: -0.0000

Data Index: 343101 (C4 (Web Text))

Max Activating Token Index: 483

Text #2

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.3385. Min Act: -0.0000

Data Index: 1362805 (Python Code)

Max Activating Token Index: 189

Text #3

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.3333. Min Act: -0.0000

Data Index: 139450 (C4 (Web Text))

Max Activating Token Index: 77

Text #4

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.3488. Min Act: -0.0000

Data Index: 1380931 (Python Code)

Max Activating Token Index: 53

Text #5

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.3397. Min Act: -0.0000

Data Index: 1435783 (Python Code)

Max Activating Token Index: 272

Text #6

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2541. Min Act: -0.0000

Data Index: 1055788 (C4 (Web Text))

Max Activating Token Index: 509

Text #7

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2155. Min Act: -0.0000

Data Index: 1563801 (Python Code)

Max Activating Token Index: 240

Text #8

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2480. Min Act: -0.0000

Data Index: 418840 (C4 (Web Text))

Max Activating Token Index: 453

Text #9

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2206. Min Act: -0.0000

Data Index: 1238616 (C4 (Web Text))

Max Activating Token Index: 638

Text #10

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2845. Min Act: -0.0000

Data Index: 147667 (C4 (Web Text))

Max Activating Token Index: 562

Text #11

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2383. Min Act: -0.0000

Data Index: 1537760 (Python Code)

Max Activating Token Index: 281

Text #12

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2818. Min Act: -0.0000

Data Index: 1414746 (Python Code)

Max Activating Token Index: 113

Text #13

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2711. Min Act: -0.0000

Data Index: 1616232 (Python Code)

Max Activating Token Index: 87

Text #14

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2245. Min Act: -0.0000

Data Index: 869012 (C4 (Web Text))

Max Activating Token Index: 740

Text #15

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.2071. Min Act: -0.0000

Data Index: 1588154 (Python Code)

Max Activating Token Index: 406

Text #16

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.1759. Min Act: -0.0000

Data Index: 1517448 (Python Code)

Max Activating Token Index: 454

Text #17

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.1602. Min Act: -0.0000

Data Index: 295931 (C4 (Web Text))

Max Activating Token Index: 787

Text #18

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.1931. Min Act: -0.0000

Data Index: 1314176 (C4 (Web Text))

Max Activating Token Index: 538

Text #19

Max Range: 1.4512. Min Range: -1.4512

Max Act: 1.1558. Min Act: -0.0000

Data Index: 1439612 (Python Code)

Max Activating Token Index: 55