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

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

Neuron 799 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: 0.0335. Min Range: -0.0335

Max Act: 0.0335. Min Act: -0.0003

Data Index: 576269 (C4 (Web Text))

Max Activating Token Index: 1021

Text #1

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0281. Min Act: -0.0003

Data Index: 206077 (C4 (Web Text))

Max Activating Token Index: 676

Text #2

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0270. Min Act: -0.0003

Data Index: 1399517 (Python Code)

Max Activating Token Index: 984

Text #3

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0282. Min Act: -0.0003

Data Index: 441883 (C4 (Web Text))

Max Activating Token Index: 762

Text #4

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0261. Min Act: -0.0003

Data Index: 312911 (C4 (Web Text))

Max Activating Token Index: 318

Text #5

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0267. Min Act: -0.0003

Data Index: 924277 (C4 (Web Text))

Max Activating Token Index: 49

Text #6

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0260. Min Act: -0.0003

Data Index: 1328553 (C4 (Web Text))

Max Activating Token Index: 497

Text #7

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0258. Min Act: -0.0002

Data Index: 979528 (C4 (Web Text))

Max Activating Token Index: 22

Text #8

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0247. Min Act: -0.0001

Data Index: 231646 (C4 (Web Text))

Max Activating Token Index: 877

Text #9

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0245. Min Act: -0.0000

Data Index: 157233 (C4 (Web Text))

Max Activating Token Index: 5

Text #10

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0244. Min Act: -0.0003

Data Index: 448833 (C4 (Web Text))

Max Activating Token Index: 161

Text #11

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0237. Min Act: -0.0001

Data Index: 1167348 (C4 (Web Text))

Max Activating Token Index: 832

Text #12

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0244. Min Act: -0.0003

Data Index: 455436 (C4 (Web Text))

Max Activating Token Index: 759

Text #13

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0244. Min Act: -0.0002

Data Index: 1359777 (C4 (Web Text))

Max Activating Token Index: 674

Text #14

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0232. Min Act: -0.0003

Data Index: 144295 (C4 (Web Text))

Max Activating Token Index: 193

Text #15

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0230. Min Act: -0.0003

Data Index: 862254 (C4 (Web Text))

Max Activating Token Index: 951

Text #16

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0237. Min Act: -0.0002

Data Index: 736491 (C4 (Web Text))

Max Activating Token Index: 535

Text #17

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0232. Min Act: -0.0003

Data Index: 1516957 (Python Code)

Max Activating Token Index: 558

Text #18

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0235. Min Act: 0.0001

Data Index: 1134902 (C4 (Web Text))

Max Activating Token Index: 259

Text #19

Max Range: 0.0335. Min Range: -0.0335

Max Act: 0.0233. Min Act: -0.0003

Data Index: 1045494 (C4 (Web Text))

Max Activating Token Index: 554