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

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

Neuron 3470 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: 0.0735. Min Range: -0.0735

Max Act: 0.0735. Min Act: -0.0000

Data Index: 571484 (C4 (Web Text))

Max Activating Token Index: 701

Text #1

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0708. Min Act: -0.0000

Data Index: 973485 (C4 (Web Text))

Max Activating Token Index: 454

Text #2

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0645. Min Act: -0.0000

Data Index: 670185 (C4 (Web Text))

Max Activating Token Index: 342

Text #3

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0630. Min Act: -0.0000

Data Index: 554733 (C4 (Web Text))

Max Activating Token Index: 805

Text #4

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0572. Min Act: -0.0000

Data Index: 953009 (C4 (Web Text))

Max Activating Token Index: 744

Text #5

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0563. Min Act: -0.0000

Data Index: 653944 (C4 (Web Text))

Max Activating Token Index: 627

Text #6

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0555. Min Act: -0.0000

Data Index: 885560 (C4 (Web Text))

Max Activating Token Index: 813

Text #7

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0543. Min Act: -0.0000

Data Index: 1620493 (Python Code)

Max Activating Token Index: 53

Text #8

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0548. Min Act: -0.0000

Data Index: 222808 (C4 (Web Text))

Max Activating Token Index: 882

Text #9

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0564. Min Act: -0.0000

Data Index: 1544415 (Python Code)

Max Activating Token Index: 456

Text #10

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0541. Min Act: -0.0000

Data Index: 1534301 (Python Code)

Max Activating Token Index: 36

Text #11

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0539. Min Act: -0.0000

Data Index: 1359277 (C4 (Web Text))

Max Activating Token Index: 291

Text #12

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0530. Min Act: -0.0000

Data Index: 11698 (C4 (Web Text))

Max Activating Token Index: 535

Text #13

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0539. Min Act: -0.0000

Data Index: 1361009 (Python Code)

Max Activating Token Index: 849

Text #14

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0567. Min Act: -0.0000

Data Index: 259314 (C4 (Web Text))

Max Activating Token Index: 221

Text #15

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0527. Min Act: -0.0000

Data Index: 746089 (C4 (Web Text))

Max Activating Token Index: 80

Text #16

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0520. Min Act: -0.0000

Data Index: 1621338 (Python Code)

Max Activating Token Index: 141

Text #17

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0520. Min Act: -0.0000

Data Index: 899288 (C4 (Web Text))

Max Activating Token Index: 469

Text #18

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0540. Min Act: -0.0000

Data Index: 584193 (C4 (Web Text))

Max Activating Token Index: 928

Text #19

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0555. Min Act: -0.0000

Data Index: 1646791 (Python Code)

Max Activating Token Index: 64