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

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

Neuron 3421 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0855. Min Act: -0.0003

Data Index: 442603 (C4 (Web Text))

Max Activating Token Index: 276

Text #1

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0801. Min Act: -0.0003

Data Index: 1413030 (Python Code)

Max Activating Token Index: 409

Text #2

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0742. Min Act: -0.0003

Data Index: 1483615 (Python Code)

Max Activating Token Index: 65

Text #3

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0802. Min Act: -0.0002

Data Index: 280806 (C4 (Web Text))

Max Activating Token Index: 652

Text #4

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0758. Min Act: -0.0003

Data Index: 1428770 (Python Code)

Max Activating Token Index: 30

Text #5

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0773. Min Act: -0.0003

Data Index: 461895 (C4 (Web Text))

Max Activating Token Index: 878

Text #6

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0727. Min Act: -0.0003

Data Index: 1609492 (Python Code)

Max Activating Token Index: 454

Text #7

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0710. Min Act: -0.0003

Data Index: 246511 (C4 (Web Text))

Max Activating Token Index: 202

Text #8

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0698. Min Act: -0.0002

Data Index: 1139128 (C4 (Web Text))

Max Activating Token Index: 290

Text #9

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0666. Min Act: -0.0003

Data Index: 77025 (C4 (Web Text))

Max Activating Token Index: 610

Text #10

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0692. Min Act: -0.0003

Data Index: 572995 (C4 (Web Text))

Max Activating Token Index: 621

Text #11

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0690. Min Act: -0.0003

Data Index: 1375999 (Python Code)

Max Activating Token Index: 228

Text #12

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0687. Min Act: -0.0003

Data Index: 1023992 (C4 (Web Text))

Max Activating Token Index: 836

Text #13

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0670. Min Act: -0.0002

Data Index: 176688 (C4 (Web Text))

Max Activating Token Index: 281

Text #14

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0674. Min Act: -0.0003

Data Index: 1440244 (Python Code)

Max Activating Token Index: 316

Text #15

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0692. Min Act: -0.0003

Data Index: 1610276 (Python Code)

Max Activating Token Index: 201

Text #16

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0659. Min Act: -0.0003

Data Index: 1467081 (Python Code)

Max Activating Token Index: 50

Text #17

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0659. Min Act: -0.0003

Data Index: 1651735 (Python Code)

Max Activating Token Index: 582

Text #18

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0653. Min Act: -0.0003

Data Index: 1628354 (Python Code)

Max Activating Token Index: 555

Text #19

Max Range: 0.0855. Min Range: -0.0855

Max Act: 0.0650. Min Act: -0.0003

Data Index: 482320 (C4 (Web Text))

Max Activating Token Index: 355