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

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

Neuron 896 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: 0.0377. Min Range: -0.0377

Max Act: 0.0363. Min Act: -0.0003

Data Index: 441241 (C4 (Web Text))

Max Activating Token Index: 667

Text #1

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0377. Min Act: -0.0004

Data Index: 334833 (C4 (Web Text))

Max Activating Token Index: 55

Text #2

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0362. Min Act: -0.0003

Data Index: 1482166 (Python Code)

Max Activating Token Index: 236

Text #3

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0356. Min Act: -0.0003

Data Index: 1442175 (Python Code)

Max Activating Token Index: 840

Text #4

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0336. Min Act: -0.0003

Data Index: 1325562 (C4 (Web Text))

Max Activating Token Index: 796

Text #5

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0342. Min Act: -0.0003

Data Index: 1024572 (C4 (Web Text))

Max Activating Token Index: 202

Text #6

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0335. Min Act: -0.0003

Data Index: 1413329 (Python Code)

Max Activating Token Index: 840

Text #7

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0332. Min Act: -0.0003

Data Index: 1265629 (C4 (Web Text))

Max Activating Token Index: 733

Text #8

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0335. Min Act: -0.0004

Data Index: 1548691 (Python Code)

Max Activating Token Index: 891

Text #9

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0329. Min Act: -0.0003

Data Index: 646520 (C4 (Web Text))

Max Activating Token Index: 388

Text #10

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0340. Min Act: -0.0004

Data Index: 541522 (C4 (Web Text))

Max Activating Token Index: 506

Text #11

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0330. Min Act: -0.0003

Data Index: 1628961 (Python Code)

Max Activating Token Index: 890

Text #12

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0326. Min Act: -0.0003

Data Index: 1251497 (C4 (Web Text))

Max Activating Token Index: 826

Text #13

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0328. Min Act: -0.0003

Data Index: 1587913 (Python Code)

Max Activating Token Index: 788

Text #14

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0330. Min Act: -0.0003

Data Index: 1329095 (C4 (Web Text))

Max Activating Token Index: 730

Text #15

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0327. Min Act: -0.0003

Data Index: 348582 (C4 (Web Text))

Max Activating Token Index: 975

Text #16

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0324. Min Act: -0.0003

Data Index: 1066655 (C4 (Web Text))

Max Activating Token Index: 32

Text #17

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0331. Min Act: -0.0003

Data Index: 1466833 (Python Code)

Max Activating Token Index: 753

Text #18

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0317. Min Act: -0.0003

Data Index: 523423 (C4 (Web Text))

Max Activating Token Index: 125

Text #19

Max Range: 0.0377. Min Range: -0.0377

Max Act: 0.0330. Min Act: -0.0003

Data Index: 1626566 (Python Code)

Max Activating Token Index: 635