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

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

Neuron 3501 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.0850. Min Range: -6.0850

Max Act: 6.0850. Min Act: -0.0000

Data Index: 518491 (C4 (Web Text))

Max Activating Token Index: 939

Text #1

Max Range: 6.0850. Min Range: -6.0850

Max Act: 5.2176. Min Act: -0.0000

Data Index: 1356537 (C4 (Web Text))

Max Activating Token Index: 587

Text #2

Max Range: 6.0850. Min Range: -6.0850

Max Act: 5.4758. Min Act: -0.0000

Data Index: 958824 (C4 (Web Text))

Max Activating Token Index: 667

Text #3

Max Range: 6.0850. Min Range: -6.0850

Max Act: 5.0803. Min Act: -0.0000

Data Index: 1000148 (C4 (Web Text))

Max Activating Token Index: 622

Text #4

Max Range: 6.0850. Min Range: -6.0850

Max Act: 5.0415. Min Act: -0.0000

Data Index: 329041 (C4 (Web Text))

Max Activating Token Index: 1003

Text #5

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.9436. Min Act: -0.0000

Data Index: 469043 (C4 (Web Text))

Max Activating Token Index: 859

Text #6

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.8695. Min Act: -0.0000

Data Index: 998585 (C4 (Web Text))

Max Activating Token Index: 696

Text #7

Max Range: 6.0850. Min Range: -6.0850

Max Act: 5.0271. Min Act: -0.0000

Data Index: 492280 (C4 (Web Text))

Max Activating Token Index: 848

Text #8

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.6080. Min Act: -0.0000

Data Index: 839705 (C4 (Web Text))

Max Activating Token Index: 667

Text #9

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.7240. Min Act: -0.0000

Data Index: 619322 (C4 (Web Text))

Max Activating Token Index: 971

Text #10

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.6599. Min Act: -0.0000

Data Index: 1140359 (C4 (Web Text))

Max Activating Token Index: 825

Text #11

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.8274. Min Act: -0.0000

Data Index: 499220 (C4 (Web Text))

Max Activating Token Index: 544

Text #12

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.5772. Min Act: -0.0000

Data Index: 222198 (C4 (Web Text))

Max Activating Token Index: 451

Text #13

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.7006. Min Act: -0.0000

Data Index: 628435 (C4 (Web Text))

Max Activating Token Index: 174

Text #14

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.4117. Min Act: -0.0000

Data Index: 324733 (C4 (Web Text))

Max Activating Token Index: 740

Text #15

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.4246. Min Act: -0.0000

Data Index: 121281 (C4 (Web Text))

Max Activating Token Index: 301

Text #16

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.3436. Min Act: -0.0000

Data Index: 1198827 (C4 (Web Text))

Max Activating Token Index: 1014

Text #17

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.4142. Min Act: -0.0000

Data Index: 1059290 (C4 (Web Text))

Max Activating Token Index: 803

Text #18

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.2308. Min Act: -0.0000

Data Index: 784873 (C4 (Web Text))

Max Activating Token Index: 979

Text #19

Max Range: 6.0850. Min Range: -6.0850

Max Act: 4.1403. Min Act: -0.0000

Data Index: 725376 (C4 (Web Text))

Max Activating Token Index: 159