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

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

Neuron 2470 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: 1.4718. Min Range: -1.4718

Max Act: 1.3981. Min Act: -0.0001

Data Index: 40879 (C4 (Web Text))

Max Activating Token Index: 694

Text #1

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.4718. Min Act: -0.0001

Data Index: 24452 (C4 (Web Text))

Max Activating Token Index: 234

Text #2

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.4222. Min Act: -0.0001

Data Index: 573975 (C4 (Web Text))

Max Activating Token Index: 691

Text #3

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.3712. Min Act: -0.0001

Data Index: 718070 (C4 (Web Text))

Max Activating Token Index: 935

Text #4

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.2254. Min Act: -0.0001

Data Index: 232880 (C4 (Web Text))

Max Activating Token Index: 156

Text #5

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1828. Min Act: -0.0001

Data Index: 622136 (C4 (Web Text))

Max Activating Token Index: 292

Text #6

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.2529. Min Act: -0.0001

Data Index: 1286081 (C4 (Web Text))

Max Activating Token Index: 136

Text #7

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1717. Min Act: -0.0001

Data Index: 497679 (C4 (Web Text))

Max Activating Token Index: 208

Text #8

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.3315. Min Act: -0.0001

Data Index: 475809 (C4 (Web Text))

Max Activating Token Index: 681

Text #9

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1384. Min Act: -0.0001

Data Index: 1301349 (C4 (Web Text))

Max Activating Token Index: 361

Text #10

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1300. Min Act: -0.0001

Data Index: 831698 (C4 (Web Text))

Max Activating Token Index: 608

Text #11

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1550. Min Act: -0.0001

Data Index: 47670 (C4 (Web Text))

Max Activating Token Index: 123

Text #12

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1231. Min Act: -0.0001

Data Index: 368324 (C4 (Web Text))

Max Activating Token Index: 674

Text #13

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.0880. Min Act: -0.0001

Data Index: 831916 (C4 (Web Text))

Max Activating Token Index: 868

Text #14

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1214. Min Act: -0.0001

Data Index: 709558 (C4 (Web Text))

Max Activating Token Index: 742

Text #15

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1196. Min Act: -0.0001

Data Index: 34371 (C4 (Web Text))

Max Activating Token Index: 837

Text #16

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1223. Min Act: -0.0001

Data Index: 794682 (C4 (Web Text))

Max Activating Token Index: 991

Text #17

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.1142. Min Act: -0.0001

Data Index: 1474790 (Python Code)

Max Activating Token Index: 237

Text #18

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.0745. Min Act: -0.0001

Data Index: 1027479 (C4 (Web Text))

Max Activating Token Index: 278

Text #19

Max Range: 1.4718. Min Range: -1.4718

Max Act: 1.0706. Min Act: -0.0001

Data Index: 797366 (C4 (Web Text))

Max Activating Token Index: 874