Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3072 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0284. Min Act: -0.0003

Data Index: 212361 (C4 (Web Text))

Max Activating Token Index: 409

Text #1

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0254. Min Act: -0.0003

Data Index: 462059 (C4 (Web Text))

Max Activating Token Index: 429

Text #2

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0246. Min Act: -0.0002

Data Index: 532552 (C4 (Web Text))

Max Activating Token Index: 325

Text #3

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0245. Min Act: -0.0003

Data Index: 501565 (C4 (Web Text))

Max Activating Token Index: 778

Text #4

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0248. Min Act: -0.0003

Data Index: 528102 (C4 (Web Text))

Max Activating Token Index: 323

Text #5

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0243. Min Act: -0.0003

Data Index: 1024695 (C4 (Web Text))

Max Activating Token Index: 482

Text #6

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0233. Min Act: -0.0003

Data Index: 521537 (C4 (Web Text))

Max Activating Token Index: 187

Text #7

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0226. Min Act: -0.0002

Data Index: 1447349 (Python Code)

Max Activating Token Index: 296

Text #8

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0221. Min Act: -0.0003

Data Index: 315493 (C4 (Web Text))

Max Activating Token Index: 469

Text #9

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0210. Min Act: -0.0002

Data Index: 275947 (C4 (Web Text))

Max Activating Token Index: 475

Text #10

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0202. Min Act: -0.0003

Data Index: 1085656 (C4 (Web Text))

Max Activating Token Index: 411

Text #11

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0210. Min Act: -0.0003

Data Index: 999719 (C4 (Web Text))

Max Activating Token Index: 615

Text #12

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0203. Min Act: -0.0003

Data Index: 84231 (C4 (Web Text))

Max Activating Token Index: 233

Text #13

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0187. Min Act: -0.0003

Data Index: 573100 (C4 (Web Text))

Max Activating Token Index: 663

Text #14

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0183. Min Act: -0.0003

Data Index: 1007055 (C4 (Web Text))

Max Activating Token Index: 333

Text #15

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0187. Min Act: -0.0003

Data Index: 49906 (C4 (Web Text))

Max Activating Token Index: 160

Text #16

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0185. Min Act: -0.0003

Data Index: 936667 (C4 (Web Text))

Max Activating Token Index: 74

Text #17

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0198. Min Act: -0.0003

Data Index: 944360 (C4 (Web Text))

Max Activating Token Index: 243

Text #18

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0184. Min Act: -0.0003

Data Index: 318990 (C4 (Web Text))

Max Activating Token Index: 836

Text #19

Max Range: 0.0284. Min Range: -0.0284

Max Act: 0.0174. Min Act: -0.0003

Data Index: 1004762 (C4 (Web Text))

Max Activating Token Index: 69