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

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

Neuron 3855 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: 5.7210. Min Range: -5.7210

Max Act: 5.5886. Min Act: -0.0001

Data Index: 363119 (C4 (Web Text))

Max Activating Token Index: 643

Text #1

Max Range: 5.7210. Min Range: -5.7210

Max Act: 5.7210. Min Act: -0.0001

Data Index: 991136 (C4 (Web Text))

Max Activating Token Index: 145

Text #2

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.2274. Min Act: -0.0001

Data Index: 916941 (C4 (Web Text))

Max Activating Token Index: 762

Text #3

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.7592. Min Act: -0.0001

Data Index: 346767 (C4 (Web Text))

Max Activating Token Index: 113

Text #4

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.9701. Min Act: -0.0001

Data Index: 206699 (C4 (Web Text))

Max Activating Token Index: 775

Text #5

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.4200. Min Act: -0.0001

Data Index: 1173562 (C4 (Web Text))

Max Activating Token Index: 575

Text #6

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.4152. Min Act: -0.0001

Data Index: 805190 (C4 (Web Text))

Max Activating Token Index: 859

Text #7

Max Range: 5.7210. Min Range: -5.7210

Max Act: 3.6296. Min Act: -0.0001

Data Index: 862986 (C4 (Web Text))

Max Activating Token Index: 304

Text #8

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.3548. Min Act: -0.0001

Data Index: 560587 (C4 (Web Text))

Max Activating Token Index: 709

Text #9

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.3444. Min Act: -0.0001

Data Index: 381757 (C4 (Web Text))

Max Activating Token Index: 152

Text #10

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.2465. Min Act: -0.0001

Data Index: 1285104 (C4 (Web Text))

Max Activating Token Index: 313

Text #11

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.0525. Min Act: -0.0001

Data Index: 1045535 (C4 (Web Text))

Max Activating Token Index: 852

Text #12

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.0564. Min Act: -0.0001

Data Index: 351744 (C4 (Web Text))

Max Activating Token Index: 83

Text #13

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.0962. Min Act: -0.0001

Data Index: 1317290 (C4 (Web Text))

Max Activating Token Index: 292

Text #14

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.2378. Min Act: -0.0001

Data Index: 312991 (C4 (Web Text))

Max Activating Token Index: 729

Text #15

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.2392. Min Act: -0.0001

Data Index: 1183927 (C4 (Web Text))

Max Activating Token Index: 870

Text #16

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.1154. Min Act: -0.0001

Data Index: 372441 (C4 (Web Text))

Max Activating Token Index: 983

Text #17

Max Range: 5.7210. Min Range: -5.7210

Max Act: 3.6254. Min Act: -0.0001

Data Index: 247119 (C4 (Web Text))

Max Activating Token Index: 315

Text #18

Max Range: 5.7210. Min Range: -5.7210

Max Act: 3.9235. Min Act: -0.0001

Data Index: 225545 (C4 (Web Text))

Max Activating Token Index: 215

Text #19

Max Range: 5.7210. Min Range: -5.7210

Max Act: 4.0261. Min Act: -0.0001

Data Index: 398409 (C4 (Web Text))

Max Activating Token Index: 435