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

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

Neuron 2008 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1627. Min Range: -3.1627

Max Act: 3.1627. Min Act: -0.0000

Data Index: 1164785 (C4 (Web Text))

Max Activating Token Index: 114

Text #1

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.6426. Min Act: -0.0000

Data Index: 1591974 (Python Code)

Max Activating Token Index: 956

Text #2

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.8761. Min Act: -0.0000

Data Index: 1604779 (Python Code)

Max Activating Token Index: 711

Text #3

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.3548. Min Act: -0.0000

Data Index: 389778 (C4 (Web Text))

Max Activating Token Index: 55

Text #4

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.4868. Min Act: -0.0000

Data Index: 1108732 (C4 (Web Text))

Max Activating Token Index: 940

Text #5

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.4818. Min Act: -0.0000

Data Index: 1404365 (Python Code)

Max Activating Token Index: 604

Text #6

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.3833. Min Act: -0.0000

Data Index: 1328288 (C4 (Web Text))

Max Activating Token Index: 352

Text #7

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.4780. Min Act: -0.0000

Data Index: 180588 (C4 (Web Text))

Max Activating Token Index: 72

Text #8

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.1078. Min Act: 0.0000

Data Index: 1066197 (C4 (Web Text))

Max Activating Token Index: 485

Text #9

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.1166. Min Act: -0.0000

Data Index: 269833 (C4 (Web Text))

Max Activating Token Index: 329

Text #10

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.1068. Min Act: -0.0000

Data Index: 585857 (C4 (Web Text))

Max Activating Token Index: 667

Text #11

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.2077. Min Act: -0.0000

Data Index: 159583 (C4 (Web Text))

Max Activating Token Index: 498

Text #12

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.2845. Min Act: -0.0000

Data Index: 417101 (C4 (Web Text))

Max Activating Token Index: 619

Text #13

Max Range: 3.1627. Min Range: -3.1627

Max Act: 1.9117. Min Act: -0.0000

Data Index: 1161152 (C4 (Web Text))

Max Activating Token Index: 366

Text #14

Max Range: 3.1627. Min Range: -3.1627

Max Act: 1.7909. Min Act: -0.0000

Data Index: 1126478 (C4 (Web Text))

Max Activating Token Index: 905

Text #15

Max Range: 3.1627. Min Range: -3.1627

Max Act: 2.1010. Min Act: -0.0000

Data Index: 636908 (C4 (Web Text))

Max Activating Token Index: 691

Text #16

Max Range: 3.1627. Min Range: -3.1627

Max Act: 1.6872. Min Act: -0.0000

Data Index: 1242194 (C4 (Web Text))

Max Activating Token Index: 647

Text #17

Max Range: 3.1627. Min Range: -3.1627

Max Act: 1.7704. Min Act: -0.0000

Data Index: 1075120 (C4 (Web Text))

Max Activating Token Index: 962

Text #18

Max Range: 3.1627. Min Range: -3.1627

Max Act: 1.6764. Min Act: -0.0000

Data Index: 1086961 (C4 (Web Text))

Max Activating Token Index: 530

Text #19

Max Range: 3.1627. Min Range: -3.1627

Max Act: 1.7516. Min Act: -0.0000

Data Index: 1502961 (Python Code)

Max Activating Token Index: 987