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

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

Neuron 355 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: 0.0444. Min Range: -0.0444

Max Act: 0.0444. Min Act: -0.0003

Data Index: 479398 (C4 (Web Text))

Max Activating Token Index: 346

Text #1

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0414. Min Act: -0.0003

Data Index: 843847 (C4 (Web Text))

Max Activating Token Index: 183

Text #2

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0396. Min Act: -0.0003

Data Index: 40106 (C4 (Web Text))

Max Activating Token Index: 89

Text #3

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0372. Min Act: -0.0003

Data Index: 1394278 (Python Code)

Max Activating Token Index: 396

Text #4

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0369. Min Act: -0.0003

Data Index: 515616 (C4 (Web Text))

Max Activating Token Index: 49

Text #5

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0375. Min Act: -0.0003

Data Index: 1042923 (C4 (Web Text))

Max Activating Token Index: 170

Text #6

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0356. Min Act: -0.0002

Data Index: 450477 (C4 (Web Text))

Max Activating Token Index: 205

Text #7

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0353. Min Act: -0.0003

Data Index: 526578 (C4 (Web Text))

Max Activating Token Index: 440

Text #8

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0349. Min Act: -0.0003

Data Index: 1017427 (C4 (Web Text))

Max Activating Token Index: 475

Text #9

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0333. Min Act: -0.0003

Data Index: 1448652 (Python Code)

Max Activating Token Index: 316

Text #10

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0350. Min Act: -0.0003

Data Index: 1495178 (Python Code)

Max Activating Token Index: 947

Text #11

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0331. Min Act: -0.0003

Data Index: 403153 (C4 (Web Text))

Max Activating Token Index: 792

Text #12

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0331. Min Act: -0.0003

Data Index: 814073 (C4 (Web Text))

Max Activating Token Index: 398

Text #13

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0314. Min Act: -0.0003

Data Index: 1551687 (Python Code)

Max Activating Token Index: 24

Text #14

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0331. Min Act: -0.0003

Data Index: 1575722 (Python Code)

Max Activating Token Index: 84

Text #15

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0329. Min Act: -0.0003

Data Index: 176645 (C4 (Web Text))

Max Activating Token Index: 867

Text #16

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0320. Min Act: -0.0003

Data Index: 212304 (C4 (Web Text))

Max Activating Token Index: 687

Text #17

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0322. Min Act: -0.0003

Data Index: 716225 (C4 (Web Text))

Max Activating Token Index: 721

Text #18

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0318. Min Act: -0.0003

Data Index: 3786 (C4 (Web Text))

Max Activating Token Index: 620

Text #19

Max Range: 0.0444. Min Range: -0.0444

Max Act: 0.0340. Min Act: -0.0003

Data Index: 1560594 (Python Code)

Max Activating Token Index: 826