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

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

Neuron 1637 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: 0.0378. Min Range: -0.0378

Max Act: 0.0378. Min Act: -0.0002

Data Index: 358000 (C4 (Web Text))

Max Activating Token Index: 1004

Text #1

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0341. Min Act: -0.0002

Data Index: 475104 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0299. Min Act: -0.0002

Data Index: 1568969 (Python Code)

Max Activating Token Index: 842

Text #3

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0296. Min Act: -0.0002

Data Index: 535995 (C4 (Web Text))

Max Activating Token Index: 864

Text #4

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0317. Min Act: -0.0002

Data Index: 1652251 (Python Code)

Max Activating Token Index: 803

Text #5

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0286. Min Act: -0.0002

Data Index: 828733 (C4 (Web Text))

Max Activating Token Index: 847

Text #6

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0309. Min Act: -0.0002

Data Index: 1487117 (Python Code)

Max Activating Token Index: 945

Text #7

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0297. Min Act: -0.0002

Data Index: 1607873 (Python Code)

Max Activating Token Index: 1008

Text #8

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0293. Min Act: -0.0002

Data Index: 1020676 (C4 (Web Text))

Max Activating Token Index: 594

Text #9

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0281. Min Act: -0.0002

Data Index: 1426717 (Python Code)

Max Activating Token Index: 819

Text #10

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0281. Min Act: -0.0003

Data Index: 1606090 (Python Code)

Max Activating Token Index: 698

Text #11

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0272. Min Act: -0.0002

Data Index: 1442324 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0270. Min Act: -0.0002

Data Index: 1553021 (Python Code)

Max Activating Token Index: 999

Text #13

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0273. Min Act: -0.0002

Data Index: 318696 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0273. Min Act: -0.0002

Data Index: 1332173 (C4 (Web Text))

Max Activating Token Index: 460

Text #15

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0268. Min Act: -0.0002

Data Index: 1387389 (Python Code)

Max Activating Token Index: 646

Text #16

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0263. Min Act: -0.0002

Data Index: 287974 (C4 (Web Text))

Max Activating Token Index: 839

Text #17

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0268. Min Act: -0.0002

Data Index: 1645480 (Python Code)

Max Activating Token Index: 596

Text #18

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0259. Min Act: -0.0002

Data Index: 352537 (C4 (Web Text))

Max Activating Token Index: 1001

Text #19

Max Range: 0.0378. Min Range: -0.0378

Max Act: 0.0258. Min Act: -0.0002

Data Index: 1652296 (Python Code)

Max Activating Token Index: 606