Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 639 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.6073. Min Act: -0.0000

Data Index: 1458815 (Python Code)

Max Activating Token Index: 898

Text #1

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.5555. Min Act: -0.0000

Data Index: 1608552 (Python Code)

Max Activating Token Index: 983

Text #2

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.2581. Min Act: -0.0000

Data Index: 1394975 (Python Code)

Max Activating Token Index: 986

Text #3

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.1517. Min Act: -0.0000

Data Index: 1504416 (Python Code)

Max Activating Token Index: 88

Text #4

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9855. Min Act: -0.0000

Data Index: 1613089 (Python Code)

Max Activating Token Index: 656

Text #5

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.1112. Min Act: -0.0000

Data Index: 90838 (C4 (Web Text))

Max Activating Token Index: 271

Text #6

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.0562. Min Act: -0.0000

Data Index: 159283 (C4 (Web Text))

Max Activating Token Index: 761

Text #7

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.0403. Min Act: -0.0000

Data Index: 1261495 (C4 (Web Text))

Max Activating Token Index: 913

Text #8

Max Range: 1.6073. Min Range: -1.6073

Max Act: 1.0651. Min Act: -0.0000

Data Index: 1381529 (Python Code)

Max Activating Token Index: 664

Text #9

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9397. Min Act: -0.0000

Data Index: 829662 (C4 (Web Text))

Max Activating Token Index: 96

Text #10

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9430. Min Act: -0.0000

Data Index: 1360245 (Python Code)

Max Activating Token Index: 202

Text #11

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9171. Min Act: -0.0000

Data Index: 731587 (C4 (Web Text))

Max Activating Token Index: 830

Text #12

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9440. Min Act: -0.0000

Data Index: 807849 (C4 (Web Text))

Max Activating Token Index: 822

Text #13

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.8925. Min Act: -0.0000

Data Index: 1652231 (Python Code)

Max Activating Token Index: 380

Text #14

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9412. Min Act: -0.0000

Data Index: 89448 (C4 (Web Text))

Max Activating Token Index: 909

Text #15

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9663. Min Act: -0.0000

Data Index: 1373124 (Python Code)

Max Activating Token Index: 559

Text #16

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.9183. Min Act: -0.0000

Data Index: 675658 (C4 (Web Text))

Max Activating Token Index: 873

Text #17

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.8675. Min Act: -0.0000

Data Index: 1626315 (Python Code)

Max Activating Token Index: 228

Text #18

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.8405. Min Act: -0.0000

Data Index: 873988 (C4 (Web Text))

Max Activating Token Index: 643

Text #19

Max Range: 1.6073. Min Range: -1.6073

Max Act: 0.8173. Min Act: -0.0000

Data Index: 1490368 (Python Code)

Max Activating Token Index: 860