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

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

Neuron 3136 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0346. Min Act: -0.0000

Data Index: 199230 (C4 (Web Text))

Max Activating Token Index: 1021

Text #1

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0340. Min Act: -0.0000

Data Index: 1504902 (Python Code)

Max Activating Token Index: 427

Text #2

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0354. Min Act: -0.0000

Data Index: 516781 (C4 (Web Text))

Max Activating Token Index: 777

Text #3

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0331. Min Act: -0.0001

Data Index: 920799 (C4 (Web Text))

Max Activating Token Index: 33

Text #4

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0301. Min Act: -0.0000

Data Index: 500950 (C4 (Web Text))

Max Activating Token Index: 446

Text #5

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0285. Min Act: -0.0000

Data Index: 211603 (C4 (Web Text))

Max Activating Token Index: 162

Text #6

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0296. Min Act: -0.0000

Data Index: 31061 (C4 (Web Text))

Max Activating Token Index: 114

Text #7

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0307. Min Act: -0.0000

Data Index: 1265267 (C4 (Web Text))

Max Activating Token Index: 821

Text #8

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0297. Min Act: -0.0000

Data Index: 706963 (C4 (Web Text))

Max Activating Token Index: 918

Text #9

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0289. Min Act: -0.0000

Data Index: 708523 (C4 (Web Text))

Max Activating Token Index: 439

Text #10

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0299. Min Act: -0.0001

Data Index: 861621 (C4 (Web Text))

Max Activating Token Index: 173

Text #11

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0311. Min Act: -0.0000

Data Index: 1110356 (C4 (Web Text))

Max Activating Token Index: 541

Text #12

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0278. Min Act: -0.0000

Data Index: 280297 (C4 (Web Text))

Max Activating Token Index: 252

Text #13

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0281. Min Act: -0.0000

Data Index: 1162821 (C4 (Web Text))

Max Activating Token Index: 852

Text #14

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0265. Min Act: -0.0001

Data Index: 1201400 (C4 (Web Text))

Max Activating Token Index: 400

Text #15

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0283. Min Act: -0.0000

Data Index: 32240 (C4 (Web Text))

Max Activating Token Index: 914

Text #16

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0247. Min Act: -0.0001

Data Index: 246998 (C4 (Web Text))

Max Activating Token Index: 568

Text #17

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0254. Min Act: -0.0000

Data Index: 606011 (C4 (Web Text))

Max Activating Token Index: 40

Text #18

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0261. Min Act: -0.0000

Data Index: 1598699 (Python Code)

Max Activating Token Index: 69

Text #19

Max Range: 0.0354. Min Range: -0.0354

Max Act: 0.0252. Min Act: -0.0000

Data Index: 1037723 (C4 (Web Text))

Max Activating Token Index: 188