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

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

Neuron 3341 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0650. Min Act: -0.0001

Data Index: 119530 (C4 (Web Text))

Max Activating Token Index: 503

Text #1

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0414. Min Act: -0.0000

Data Index: 1428482 (Python Code)

Max Activating Token Index: 144

Text #2

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0458. Min Act: -0.0001

Data Index: 1263848 (C4 (Web Text))

Max Activating Token Index: 677

Text #3

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0397. Min Act: -0.0001

Data Index: 1275446 (C4 (Web Text))

Max Activating Token Index: 464

Text #4

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0355. Min Act: -0.0001

Data Index: 1140720 (C4 (Web Text))

Max Activating Token Index: 139

Text #5

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0377. Min Act: -0.0001

Data Index: 1529724 (Python Code)

Max Activating Token Index: 516

Text #6

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0374. Min Act: -0.0001

Data Index: 106729 (C4 (Web Text))

Max Activating Token Index: 536

Text #7

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0352. Min Act: -0.0001

Data Index: 1008857 (C4 (Web Text))

Max Activating Token Index: 926

Text #8

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0313. Min Act: -0.0001

Data Index: 584543 (C4 (Web Text))

Max Activating Token Index: 320

Text #9

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0400. Min Act: -0.0001

Data Index: 1234758 (C4 (Web Text))

Max Activating Token Index: 546

Text #10

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0343. Min Act: -0.0001

Data Index: 1597764 (Python Code)

Max Activating Token Index: 190

Text #11

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0357. Min Act: -0.0000

Data Index: 279277 (C4 (Web Text))

Max Activating Token Index: 621

Text #12

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0318. Min Act: -0.0001

Data Index: 750500 (C4 (Web Text))

Max Activating Token Index: 734

Text #13

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0309. Min Act: -0.0001

Data Index: 1132231 (C4 (Web Text))

Max Activating Token Index: 675

Text #14

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0338. Min Act: -0.0001

Data Index: 1020631 (C4 (Web Text))

Max Activating Token Index: 232

Text #15

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0333. Min Act: -0.0001

Data Index: 1656024 (Python Code)

Max Activating Token Index: 220

Text #16

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0321. Min Act: -0.0001

Data Index: 870228 (C4 (Web Text))

Max Activating Token Index: 694

Text #17

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0353. Min Act: -0.0001

Data Index: 1456266 (Python Code)

Max Activating Token Index: 183

Text #18

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0349. Min Act: -0.0001

Data Index: 1465852 (Python Code)

Max Activating Token Index: 597

Text #19

Max Range: 0.0650. Min Range: -0.0650

Max Act: 0.0302. Min Act: -0.0001

Data Index: 140446 (C4 (Web Text))

Max Activating Token Index: 486