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

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

Neuron 2016 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0436. Min Act: -0.0006

Data Index: 1102543 (C4 (Web Text))

Max Activating Token Index: 837

Text #1

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0317. Min Act: -0.0006

Data Index: 255695 (C4 (Web Text))

Max Activating Token Index: 768

Text #2

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0309. Min Act: -0.0006

Data Index: 209577 (C4 (Web Text))

Max Activating Token Index: 917

Text #3

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0301. Min Act: -0.0006

Data Index: 1561214 (Python Code)

Max Activating Token Index: 953

Text #4

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0287. Min Act: -0.0006

Data Index: 678953 (C4 (Web Text))

Max Activating Token Index: 763

Text #5

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0286. Min Act: -0.0006

Data Index: 109571 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0274. Min Act: -0.0006

Data Index: 1010726 (C4 (Web Text))

Max Activating Token Index: 719

Text #7

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0282. Min Act: -0.0006

Data Index: 1226800 (C4 (Web Text))

Max Activating Token Index: 1015

Text #8

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0288. Min Act: -0.0006

Data Index: 507646 (C4 (Web Text))

Max Activating Token Index: 722

Text #9

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0275. Min Act: -0.0006

Data Index: 126372 (C4 (Web Text))

Max Activating Token Index: 957

Text #10

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0274. Min Act: -0.0006

Data Index: 473317 (C4 (Web Text))

Max Activating Token Index: 686

Text #11

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0267. Min Act: -0.0006

Data Index: 1228780 (C4 (Web Text))

Max Activating Token Index: 599

Text #12

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0267. Min Act: -0.0006

Data Index: 1170985 (C4 (Web Text))

Max Activating Token Index: 825

Text #13

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0272. Min Act: -0.0006

Data Index: 1578880 (Python Code)

Max Activating Token Index: 613

Text #14

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0260. Min Act: -0.0006

Data Index: 1425015 (Python Code)

Max Activating Token Index: 440

Text #15

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0269. Min Act: -0.0006

Data Index: 1279766 (C4 (Web Text))

Max Activating Token Index: 749

Text #16

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0262. Min Act: -0.0006

Data Index: 516919 (C4 (Web Text))

Max Activating Token Index: 836

Text #17

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0255. Min Act: -0.0006

Data Index: 1337698 (C4 (Web Text))

Max Activating Token Index: 951

Text #18

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0260. Min Act: -0.0006

Data Index: 427317 (C4 (Web Text))

Max Activating Token Index: 940

Text #19

Max Range: 0.0436. Min Range: -0.0436

Max Act: 0.0259. Min Act: -0.0006

Data Index: 1029251 (C4 (Web Text))

Max Activating Token Index: 761