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

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

Neuron 651 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1714. Min Range: -1.1714

Max Act: 1.1714. Min Act: -0.0001

Data Index: 481886 (C4 (Web Text))

Max Activating Token Index: 899

Text #1

Max Range: 1.1714. Min Range: -1.1714

Max Act: 1.1363. Min Act: -0.0001

Data Index: 1419457 (Python Code)

Max Activating Token Index: 652

Text #2

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9920. Min Act: -0.0001

Data Index: 885147 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 1.1714. Min Range: -1.1714

Max Act: 1.0666. Min Act: -0.0001

Data Index: 307427 (C4 (Web Text))

Max Activating Token Index: 932

Text #4

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9778. Min Act: -0.0001

Data Index: 686910 (C4 (Web Text))

Max Activating Token Index: 206

Text #5

Max Range: 1.1714. Min Range: -1.1714

Max Act: 1.0101. Min Act: -0.0001

Data Index: 868193 (C4 (Web Text))

Max Activating Token Index: 434

Text #6

Max Range: 1.1714. Min Range: -1.1714

Max Act: 1.0140. Min Act: -0.0001

Data Index: 810064 (C4 (Web Text))

Max Activating Token Index: 162

Text #7

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.8934. Min Act: -0.0001

Data Index: 1608883 (Python Code)

Max Activating Token Index: 910

Text #8

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9198. Min Act: -0.0001

Data Index: 209854 (C4 (Web Text))

Max Activating Token Index: 17

Text #9

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9791. Min Act: -0.0001

Data Index: 251327 (C4 (Web Text))

Max Activating Token Index: 612

Text #10

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.8967. Min Act: -0.0001

Data Index: 119311 (C4 (Web Text))

Max Activating Token Index: 785

Text #11

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9232. Min Act: -0.0001

Data Index: 147666 (C4 (Web Text))

Max Activating Token Index: 849

Text #12

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.8091. Min Act: -0.0001

Data Index: 394869 (C4 (Web Text))

Max Activating Token Index: 276

Text #13

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9036. Min Act: -0.0001

Data Index: 1478836 (Python Code)

Max Activating Token Index: 642

Text #14

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.7406. Min Act: -0.0001

Data Index: 324883 (C4 (Web Text))

Max Activating Token Index: 545

Text #15

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9263. Min Act: -0.0001

Data Index: 1301381 (C4 (Web Text))

Max Activating Token Index: 868

Text #16

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.8499. Min Act: -0.0001

Data Index: 372891 (C4 (Web Text))

Max Activating Token Index: 726

Text #17

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.8795. Min Act: -0.0001

Data Index: 160472 (C4 (Web Text))

Max Activating Token Index: 898

Text #18

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.9067. Min Act: -0.0001

Data Index: 620412 (C4 (Web Text))

Max Activating Token Index: 932

Text #19

Max Range: 1.1714. Min Range: -1.1714

Max Act: 0.8328. Min Act: -0.0001

Data Index: 1492680 (Python Code)

Max Activating Token Index: 626