Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1551 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0393. Min Act: -0.0003

Data Index: 92847 (C4 (Web Text))

Max Activating Token Index: 831

Text #1

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0304. Min Act: -0.0002

Data Index: 1427187 (Python Code)

Max Activating Token Index: 676

Text #2

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0310. Min Act: -0.0003

Data Index: 1596884 (Python Code)

Max Activating Token Index: 407

Text #3

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0304. Min Act: -0.0002

Data Index: 29779 (C4 (Web Text))

Max Activating Token Index: 671

Text #4

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0287. Min Act: -0.0003

Data Index: 955436 (C4 (Web Text))

Max Activating Token Index: 476

Text #5

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0292. Min Act: -0.0002

Data Index: 308619 (C4 (Web Text))

Max Activating Token Index: 866

Text #6

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0296. Min Act: -0.0002

Data Index: 1646869 (Python Code)

Max Activating Token Index: 945

Text #7

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0289. Min Act: -0.0002

Data Index: 686464 (C4 (Web Text))

Max Activating Token Index: 686

Text #8

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0281. Min Act: -0.0003

Data Index: 1071914 (C4 (Web Text))

Max Activating Token Index: 481

Text #9

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0287. Min Act: -0.0003

Data Index: 1295101 (C4 (Web Text))

Max Activating Token Index: 642

Text #10

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0285. Min Act: -0.0003

Data Index: 463196 (C4 (Web Text))

Max Activating Token Index: 223

Text #11

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0284. Min Act: -0.0004

Data Index: 215583 (C4 (Web Text))

Max Activating Token Index: 892

Text #12

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0292. Min Act: -0.0003

Data Index: 420126 (C4 (Web Text))

Max Activating Token Index: 295

Text #13

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0286. Min Act: -0.0002

Data Index: 1540003 (Python Code)

Max Activating Token Index: 817

Text #14

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0282. Min Act: -0.0003

Data Index: 1443859 (Python Code)

Max Activating Token Index: 497

Text #15

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0284. Min Act: -0.0002

Data Index: 1552292 (Python Code)

Max Activating Token Index: 675

Text #16

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0268. Min Act: -0.0002

Data Index: 731594 (C4 (Web Text))

Max Activating Token Index: 890

Text #17

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0270. Min Act: -0.0003

Data Index: 1548825 (Python Code)

Max Activating Token Index: 61

Text #18

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0263. Min Act: -0.0002

Data Index: 1530170 (Python Code)

Max Activating Token Index: 882

Text #19

Max Range: 0.0393. Min Range: -0.0393

Max Act: 0.0265. Min Act: -0.0002

Data Index: 1570057 (Python Code)

Max Activating Token Index: 777