Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1794 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0370. Min Act: -0.0001

Data Index: 610245 (C4 (Web Text))

Max Activating Token Index: 2

Text #1

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0369. Min Act: -0.0001

Data Index: 1605310 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0351. Min Act: -0.0002

Data Index: 1099453 (C4 (Web Text))

Max Activating Token Index: 976

Text #3

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0332. Min Act: -0.0002

Data Index: 700721 (C4 (Web Text))

Max Activating Token Index: 185

Text #4

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0194. Min Act: -0.0001

Data Index: 179334 (C4 (Web Text))

Max Activating Token Index: 950

Text #5

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0325. Min Act: -0.0002

Data Index: 1060193 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0294. Min Act: -0.0001

Data Index: 468054 (C4 (Web Text))

Max Activating Token Index: 70

Text #7

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0332. Min Act: -0.0002

Data Index: 565054 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0329. Min Act: -0.0001

Data Index: 530436 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0311. Min Act: -0.0001

Data Index: 522674 (C4 (Web Text))

Max Activating Token Index: 255

Text #10

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0295. Min Act: -0.0002

Data Index: 851479 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0295. Min Act: -0.0001

Data Index: 751418 (C4 (Web Text))

Max Activating Token Index: 162

Text #12

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0300. Min Act: -0.0002

Data Index: 361371 (C4 (Web Text))

Max Activating Token Index: 504

Text #13

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0261. Min Act: -0.0002

Data Index: 1069391 (C4 (Web Text))

Max Activating Token Index: 298

Text #14

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0287. Min Act: -0.0001

Data Index: 201263 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0280. Min Act: -0.0001

Data Index: 980806 (C4 (Web Text))

Max Activating Token Index: 226

Text #16

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0268. Min Act: -0.0001

Data Index: 1324488 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0286. Min Act: -0.0001

Data Index: 1519879 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0264. Min Act: -0.0001

Data Index: 1462521 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 0.0370. Min Range: -0.0370

Max Act: 0.0273. Min Act: -0.0001

Data Index: 149693 (C4 (Web Text))

Max Activating Token Index: 867