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

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

Neuron 3435 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.7640. Min Act: -0.0000

Data Index: 1629237 (Python Code)

Max Activating Token Index: 559

Text #1

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.6620. Min Act: -0.0000

Data Index: 1541995 (Python Code)

Max Activating Token Index: 620

Text #2

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.7782. Min Act: -0.0000

Data Index: 1592658 (Python Code)

Max Activating Token Index: 207

Text #3

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.7477. Min Act: -0.0000

Data Index: 146987 (C4 (Web Text))

Max Activating Token Index: 937

Text #4

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.8281. Min Act: -0.0000

Data Index: 648835 (C4 (Web Text))

Max Activating Token Index: 208

Text #5

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.8511. Min Act: -0.0000

Data Index: 1425312 (Python Code)

Max Activating Token Index: 194

Text #6

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.6257. Min Act: -0.0000

Data Index: 27963 (C4 (Web Text))

Max Activating Token Index: 899

Text #7

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.6719. Min Act: -0.0000

Data Index: 690954 (C4 (Web Text))

Max Activating Token Index: 558

Text #8

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.6275. Min Act: -0.0000

Data Index: 1315322 (C4 (Web Text))

Max Activating Token Index: 869

Text #9

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.6097. Min Act: -0.0000

Data Index: 419714 (C4 (Web Text))

Max Activating Token Index: 545

Text #10

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.5225. Min Act: -0.0000

Data Index: 973219 (C4 (Web Text))

Max Activating Token Index: 16

Text #11

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.5294. Min Act: -0.0000

Data Index: 583980 (C4 (Web Text))

Max Activating Token Index: 32

Text #12

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.5018. Min Act: -0.0000

Data Index: 1067646 (C4 (Web Text))

Max Activating Token Index: 570

Text #13

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.5591. Min Act: -0.0000

Data Index: 523401 (C4 (Web Text))

Max Activating Token Index: 982

Text #14

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.4256. Min Act: -0.0000

Data Index: 8442 (C4 (Web Text))

Max Activating Token Index: 80

Text #15

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.5611. Min Act: -0.0000

Data Index: 1431793 (Python Code)

Max Activating Token Index: 995

Text #16

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.4104. Min Act: -0.0000

Data Index: 1296321 (C4 (Web Text))

Max Activating Token Index: 57

Text #17

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.3044. Min Act: -0.0000

Data Index: 412811 (C4 (Web Text))

Max Activating Token Index: 63

Text #18

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.3667. Min Act: -0.0000

Data Index: 497615 (C4 (Web Text))

Max Activating Token Index: 50

Text #19

Max Range: 1.8511. Min Range: -1.8511

Max Act: 1.3520. Min Act: -0.0000

Data Index: 899469 (C4 (Web Text))

Max Activating Token Index: 313