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

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

Neuron 5075 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: 5.2001. Min Range: -5.2001

Max Act: 5.2001. Min Act: -0.0000

Data Index: 184666 (C4 (Web Text))

Max Activating Token Index: 275

Text #1

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.5733. Min Act: -0.0000

Data Index: 709860 (C4 (Web Text))

Max Activating Token Index: 349

Text #2

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.9195. Min Act: -0.0000

Data Index: 409129 (C4 (Web Text))

Max Activating Token Index: 185

Text #3

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.4828. Min Act: -0.0000

Data Index: 49502 (C4 (Web Text))

Max Activating Token Index: 113

Text #4

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.5098. Min Act: -0.0000

Data Index: 562322 (C4 (Web Text))

Max Activating Token Index: 71

Text #5

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.4427. Min Act: -0.0000

Data Index: 1342492 (C4 (Web Text))

Max Activating Token Index: 893

Text #6

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.4292. Min Act: -0.0000

Data Index: 883854 (C4 (Web Text))

Max Activating Token Index: 675

Text #7

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.2376. Min Act: -0.0000

Data Index: 1253757 (C4 (Web Text))

Max Activating Token Index: 413

Text #8

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.3303. Min Act: -0.0000

Data Index: 712836 (C4 (Web Text))

Max Activating Token Index: 210

Text #9

Max Range: 5.2001. Min Range: -5.2001

Max Act: 3.9380. Min Act: -0.0000

Data Index: 1535435 (Python Code)

Max Activating Token Index: 973

Text #10

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.2816. Min Act: -0.0000

Data Index: 984938 (C4 (Web Text))

Max Activating Token Index: 596

Text #11

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.4601. Min Act: -0.0000

Data Index: 194352 (C4 (Web Text))

Max Activating Token Index: 65

Text #12

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.2711. Min Act: -0.0000

Data Index: 289449 (C4 (Web Text))

Max Activating Token Index: 68

Text #13

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.2638. Min Act: -0.0000

Data Index: 600812 (C4 (Web Text))

Max Activating Token Index: 812

Text #14

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.5350. Min Act: -0.0000

Data Index: 601133 (C4 (Web Text))

Max Activating Token Index: 823

Text #15

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.1801. Min Act: -0.0000

Data Index: 810479 (C4 (Web Text))

Max Activating Token Index: 550

Text #16

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.1602. Min Act: -0.0000

Data Index: 1431884 (Python Code)

Max Activating Token Index: 576

Text #17

Max Range: 5.2001. Min Range: -5.2001

Max Act: 3.9690. Min Act: -0.0000

Data Index: 963950 (C4 (Web Text))

Max Activating Token Index: 875

Text #18

Max Range: 5.2001. Min Range: -5.2001

Max Act: 3.9781. Min Act: -0.0000

Data Index: 400509 (C4 (Web Text))

Max Activating Token Index: 547

Text #19

Max Range: 5.2001. Min Range: -5.2001

Max Act: 4.1121. Min Act: -0.0000

Data Index: 997295 (C4 (Web Text))

Max Activating Token Index: 527