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

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

Neuron 634 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: 0.0507. Min Range: -0.0507

Max Act: 0.0507. Min Act: -0.0003

Data Index: 468820 (C4 (Web Text))

Max Activating Token Index: 335

Text #1

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0415. Min Act: -0.0002

Data Index: 618270 (C4 (Web Text))

Max Activating Token Index: 357

Text #2

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0350. Min Act: -0.0004

Data Index: 1113008 (C4 (Web Text))

Max Activating Token Index: 983

Text #3

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0326. Min Act: -0.0003

Data Index: 481173 (C4 (Web Text))

Max Activating Token Index: 574

Text #4

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0323. Min Act: -0.0003

Data Index: 415601 (C4 (Web Text))

Max Activating Token Index: 356

Text #5

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0317. Min Act: -0.0002

Data Index: 531497 (C4 (Web Text))

Max Activating Token Index: 630

Text #6

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0308. Min Act: -0.0003

Data Index: 88101 (C4 (Web Text))

Max Activating Token Index: 1005

Text #7

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0301. Min Act: -0.0003

Data Index: 870055 (C4 (Web Text))

Max Activating Token Index: 373

Text #8

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0299. Min Act: -0.0003

Data Index: 1176675 (C4 (Web Text))

Max Activating Token Index: 1000

Text #9

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0291. Min Act: -0.0003

Data Index: 900862 (C4 (Web Text))

Max Activating Token Index: 301

Text #10

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0304. Min Act: -0.0003

Data Index: 570687 (C4 (Web Text))

Max Activating Token Index: 979

Text #11

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0289. Min Act: -0.0003

Data Index: 192653 (C4 (Web Text))

Max Activating Token Index: 296

Text #12

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0290. Min Act: -0.0004

Data Index: 233026 (C4 (Web Text))

Max Activating Token Index: 1002

Text #13

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0298. Min Act: -0.0003

Data Index: 720479 (C4 (Web Text))

Max Activating Token Index: 341

Text #14

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0305. Min Act: -0.0003

Data Index: 333214 (C4 (Web Text))

Max Activating Token Index: 294

Text #15

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0288. Min Act: -0.0004

Data Index: 1310925 (C4 (Web Text))

Max Activating Token Index: 968

Text #16

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0298. Min Act: -0.0004

Data Index: 803675 (C4 (Web Text))

Max Activating Token Index: 867

Text #17

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0284. Min Act: -0.0003

Data Index: 782125 (C4 (Web Text))

Max Activating Token Index: 281

Text #18

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0295. Min Act: -0.0003

Data Index: 305963 (C4 (Web Text))

Max Activating Token Index: 376

Text #19

Max Range: 0.0507. Min Range: -0.0507

Max Act: 0.0286. Min Act: -0.0004

Data Index: 983575 (C4 (Web Text))

Max Activating Token Index: 413