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

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

Neuron 140 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: 0.0322. Min Range: -0.0322

Max Act: 0.0322. Min Act: -0.0002

Data Index: 998845 (C4 (Web Text))

Max Activating Token Index: 57

Text #1

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0300. Min Act: -0.0002

Data Index: 735367 (C4 (Web Text))

Max Activating Token Index: 32

Text #2

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0293. Min Act: -0.0003

Data Index: 601178 (C4 (Web Text))

Max Activating Token Index: 910

Text #3

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0277. Min Act: -0.0002

Data Index: 615778 (C4 (Web Text))

Max Activating Token Index: 198

Text #4

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0272. Min Act: -0.0002

Data Index: 1022601 (C4 (Web Text))

Max Activating Token Index: 667

Text #5

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0266. Min Act: -0.0003

Data Index: 801249 (C4 (Web Text))

Max Activating Token Index: 242

Text #6

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0262. Min Act: -0.0002

Data Index: 378571 (C4 (Web Text))

Max Activating Token Index: 204

Text #7

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0255. Min Act: -0.0002

Data Index: 1301449 (C4 (Web Text))

Max Activating Token Index: 670

Text #8

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0260. Min Act: -0.0003

Data Index: 1330203 (C4 (Web Text))

Max Activating Token Index: 131

Text #9

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0260. Min Act: -0.0002

Data Index: 151244 (C4 (Web Text))

Max Activating Token Index: 693

Text #10

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0244. Min Act: -0.0003

Data Index: 360334 (C4 (Web Text))

Max Activating Token Index: 264

Text #11

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0241. Min Act: -0.0002

Data Index: 692080 (C4 (Web Text))

Max Activating Token Index: 53

Text #12

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0240. Min Act: -0.0002

Data Index: 1567762 (Python Code)

Max Activating Token Index: 233

Text #13

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0242. Min Act: -0.0003

Data Index: 203117 (C4 (Web Text))

Max Activating Token Index: 901

Text #14

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0241. Min Act: -0.0003

Data Index: 310889 (C4 (Web Text))

Max Activating Token Index: 386

Text #15

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0249. Min Act: -0.0002

Data Index: 1313518 (C4 (Web Text))

Max Activating Token Index: 604

Text #16

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0238. Min Act: -0.0003

Data Index: 1281563 (C4 (Web Text))

Max Activating Token Index: 663

Text #17

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0239. Min Act: -0.0003

Data Index: 1219358 (C4 (Web Text))

Max Activating Token Index: 439

Text #18

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0227. Min Act: -0.0003

Data Index: 1108294 (C4 (Web Text))

Max Activating Token Index: 27

Text #19

Max Range: 0.0322. Min Range: -0.0322

Max Act: 0.0232. Min Act: -0.0003

Data Index: 82462 (C4 (Web Text))

Max Activating Token Index: 264