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

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

Neuron 558 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0291. Min Act: -0.0006

Data Index: 120912 (C4 (Web Text))

Max Activating Token Index: 55

Text #1

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0249. Min Act: -0.0006

Data Index: 132452 (C4 (Web Text))

Max Activating Token Index: 432

Text #2

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0243. Min Act: -0.0006

Data Index: 969570 (C4 (Web Text))

Max Activating Token Index: 370

Text #3

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0232. Min Act: -0.0006

Data Index: 1006836 (C4 (Web Text))

Max Activating Token Index: 391

Text #4

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0220. Min Act: -0.0006

Data Index: 50577 (C4 (Web Text))

Max Activating Token Index: 453

Text #5

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0219. Min Act: -0.0006

Data Index: 416055 (C4 (Web Text))

Max Activating Token Index: 603

Text #6

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0215. Min Act: -0.0006

Data Index: 646657 (C4 (Web Text))

Max Activating Token Index: 59

Text #7

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0213. Min Act: -0.0006

Data Index: 894375 (C4 (Web Text))

Max Activating Token Index: 690

Text #8

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0213. Min Act: -0.0006

Data Index: 1319033 (C4 (Web Text))

Max Activating Token Index: 18

Text #9

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0204. Min Act: -0.0006

Data Index: 934360 (C4 (Web Text))

Max Activating Token Index: 556

Text #10

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0208. Min Act: -0.0006

Data Index: 608697 (C4 (Web Text))

Max Activating Token Index: 775

Text #11

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0207. Min Act: -0.0006

Data Index: 408296 (C4 (Web Text))

Max Activating Token Index: 285

Text #12

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0207. Min Act: -0.0006

Data Index: 392826 (C4 (Web Text))

Max Activating Token Index: 24

Text #13

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0205. Min Act: -0.0006

Data Index: 1242281 (C4 (Web Text))

Max Activating Token Index: 47

Text #14

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0209. Min Act: -0.0006

Data Index: 39574 (C4 (Web Text))

Max Activating Token Index: 41

Text #15

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0200. Min Act: -0.0006

Data Index: 1260235 (C4 (Web Text))

Max Activating Token Index: 49

Text #16

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0200. Min Act: -0.0005

Data Index: 123521 (C4 (Web Text))

Max Activating Token Index: 80

Text #17

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0202. Min Act: -0.0006

Data Index: 63165 (C4 (Web Text))

Max Activating Token Index: 44

Text #18

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0201. Min Act: -0.0006

Data Index: 774602 (C4 (Web Text))

Max Activating Token Index: 171

Text #19

Max Range: 0.0291. Min Range: -0.0291

Max Act: 0.0201. Min Act: -0.0006

Data Index: 1189847 (C4 (Web Text))

Max Activating Token Index: 405