Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4750 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.1561. Min Act: -0.0000

Data Index: 1298506 (C4 (Web Text))

Max Activating Token Index: 66

Text #1

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.1632. Min Act: -0.0001

Data Index: 697959 (C4 (Web Text))

Max Activating Token Index: 189

Text #2

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.1350. Min Act: -0.0000

Data Index: 716810 (C4 (Web Text))

Max Activating Token Index: 196

Text #3

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.1058. Min Act: -0.0000

Data Index: 395679 (C4 (Web Text))

Max Activating Token Index: 777

Text #4

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.1086. Min Act: -0.0000

Data Index: 305771 (C4 (Web Text))

Max Activating Token Index: 1013

Text #5

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.1016. Min Act: -0.0000

Data Index: 1172719 (C4 (Web Text))

Max Activating Token Index: 181

Text #6

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0949. Min Act: -0.0000

Data Index: 231922 (C4 (Web Text))

Max Activating Token Index: 412

Text #7

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0948. Min Act: -0.0000

Data Index: 259379 (C4 (Web Text))

Max Activating Token Index: 84

Text #8

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0941. Min Act: -0.0000

Data Index: 605774 (C4 (Web Text))

Max Activating Token Index: 495

Text #9

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0904. Min Act: -0.0000

Data Index: 759497 (C4 (Web Text))

Max Activating Token Index: 593

Text #10

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0859. Min Act: -0.0000

Data Index: 327375 (C4 (Web Text))

Max Activating Token Index: 239

Text #11

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0876. Min Act: -0.0000

Data Index: 924320 (C4 (Web Text))

Max Activating Token Index: 720

Text #12

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0850. Min Act: -0.0000

Data Index: 837922 (C4 (Web Text))

Max Activating Token Index: 870

Text #13

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0894. Min Act: -0.0000

Data Index: 109552 (C4 (Web Text))

Max Activating Token Index: 501

Text #14

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0882. Min Act: -0.0000

Data Index: 1363893 (Python Code)

Max Activating Token Index: 547

Text #15

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0836. Min Act: -0.0000

Data Index: 29376 (C4 (Web Text))

Max Activating Token Index: 649

Text #16

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0853. Min Act: -0.0000

Data Index: 1277463 (C4 (Web Text))

Max Activating Token Index: 869

Text #17

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0765. Min Act: -0.0001

Data Index: 480879 (C4 (Web Text))

Max Activating Token Index: 633

Text #18

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0796. Min Act: -0.0000

Data Index: 318096 (C4 (Web Text))

Max Activating Token Index: 256

Text #19

Max Range: 0.1632. Min Range: -0.1632

Max Act: 0.0755. Min Act: -0.0000

Data Index: 455699 (C4 (Web Text))

Max Activating Token Index: 1010