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

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

Neuron 4223 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5344. Min Range: -2.5344

Max Act: 2.5344. Min Act: -0.0001

Data Index: 1567283 (Python Code)

Max Activating Token Index: 162

Text #1

Max Range: 2.5344. Min Range: -2.5344

Max Act: 2.1718. Min Act: -0.0000

Data Index: 860934 (C4 (Web Text))

Max Activating Token Index: 1021

Text #2

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.9935. Min Act: -0.0000

Data Index: 573785 (C4 (Web Text))

Max Activating Token Index: 869

Text #3

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.8657. Min Act: -0.0000

Data Index: 1561942 (Python Code)

Max Activating Token Index: 296

Text #4

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.7706. Min Act: -0.0000

Data Index: 746406 (C4 (Web Text))

Max Activating Token Index: 895

Text #5

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.8641. Min Act: -0.0000

Data Index: 864600 (C4 (Web Text))

Max Activating Token Index: 704

Text #6

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.6124. Min Act: -0.0000

Data Index: 884657 (C4 (Web Text))

Max Activating Token Index: 814

Text #7

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.7894. Min Act: -0.0000

Data Index: 64605 (C4 (Web Text))

Max Activating Token Index: 744

Text #8

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.5979. Min Act: -0.0000

Data Index: 1633488 (Python Code)

Max Activating Token Index: 1022

Text #9

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.5778. Min Act: -0.0000

Data Index: 563712 (C4 (Web Text))

Max Activating Token Index: 744

Text #10

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.6563. Min Act: -0.0000

Data Index: 283466 (C4 (Web Text))

Max Activating Token Index: 338

Text #11

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.7223. Min Act: -0.0001

Data Index: 911919 (C4 (Web Text))

Max Activating Token Index: 307

Text #12

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.5837. Min Act: -0.0000

Data Index: 1391426 (Python Code)

Max Activating Token Index: 66

Text #13

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.6580. Min Act: -0.0000

Data Index: 29064 (C4 (Web Text))

Max Activating Token Index: 879

Text #14

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.4215. Min Act: -0.0000

Data Index: 695628 (C4 (Web Text))

Max Activating Token Index: 682

Text #15

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.4937. Min Act: -0.0000

Data Index: 220985 (C4 (Web Text))

Max Activating Token Index: 463

Text #16

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.4629. Min Act: -0.0000

Data Index: 1175331 (C4 (Web Text))

Max Activating Token Index: 156

Text #17

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.4447. Min Act: -0.0000

Data Index: 981383 (C4 (Web Text))

Max Activating Token Index: 676

Text #18

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.3376. Min Act: -0.0000

Data Index: 512232 (C4 (Web Text))

Max Activating Token Index: 118

Text #19

Max Range: 2.5344. Min Range: -2.5344

Max Act: 1.4939. Min Act: -0.0000

Data Index: 1027989 (C4 (Web Text))

Max Activating Token Index: 71