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

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

Neuron 768 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.2554. Min Act: -0.0001

Data Index: 1436518 (Python Code)

Max Activating Token Index: 19

Text #1

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1888. Min Act: -0.0001

Data Index: 1313570 (C4 (Web Text))

Max Activating Token Index: 597

Text #2

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1761. Min Act: -0.0001

Data Index: 921665 (C4 (Web Text))

Max Activating Token Index: 58

Text #3

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1760. Min Act: -0.0001

Data Index: 913715 (C4 (Web Text))

Max Activating Token Index: 1002

Text #4

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1733. Min Act: -0.0001

Data Index: 1614770 (Python Code)

Max Activating Token Index: 12

Text #5

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1642. Min Act: -0.0001

Data Index: 1070599 (C4 (Web Text))

Max Activating Token Index: 159

Text #6

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1725. Min Act: -0.0001

Data Index: 538606 (C4 (Web Text))

Max Activating Token Index: 615

Text #7

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1544. Min Act: -0.0001

Data Index: 1220493 (C4 (Web Text))

Max Activating Token Index: 909

Text #8

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1619. Min Act: -0.0001

Data Index: 616150 (C4 (Web Text))

Max Activating Token Index: 702

Text #9

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1645. Min Act: -0.0001

Data Index: 1369036 (Python Code)

Max Activating Token Index: 126

Text #10

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1613. Min Act: -0.0001

Data Index: 399496 (C4 (Web Text))

Max Activating Token Index: 38

Text #11

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1552. Min Act: -0.0001

Data Index: 1405397 (Python Code)

Max Activating Token Index: 135

Text #12

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1399. Min Act: -0.0001

Data Index: 387656 (C4 (Web Text))

Max Activating Token Index: 673

Text #13

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1715. Min Act: -0.0001

Data Index: 1442325 (Python Code)

Max Activating Token Index: 675

Text #14

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1509. Min Act: -0.0000

Data Index: 705619 (C4 (Web Text))

Max Activating Token Index: 527

Text #15

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1339. Min Act: -0.0001

Data Index: 350694 (C4 (Web Text))

Max Activating Token Index: 36

Text #16

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1345. Min Act: -0.0001

Data Index: 591041 (C4 (Web Text))

Max Activating Token Index: 626

Text #17

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1458. Min Act: -0.0001

Data Index: 1305726 (C4 (Web Text))

Max Activating Token Index: 698

Text #18

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1393. Min Act: -0.0001

Data Index: 339360 (C4 (Web Text))

Max Activating Token Index: 528

Text #19

Max Range: 0.2554. Min Range: -0.2554

Max Act: 0.1398. Min Act: -0.0001

Data Index: 1394224 (Python Code)

Max Activating Token Index: 686