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

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

Neuron 3692 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0891. Min Act: -0.0002

Data Index: 691271 (C4 (Web Text))

Max Activating Token Index: 26

Text #1

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0825. Min Act: -0.0002

Data Index: 941584 (C4 (Web Text))

Max Activating Token Index: 25

Text #2

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0820. Min Act: -0.0002

Data Index: 579505 (C4 (Web Text))

Max Activating Token Index: 4

Text #3

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0794. Min Act: -0.0002

Data Index: 1513997 (Python Code)

Max Activating Token Index: 92

Text #4

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0819. Min Act: -0.0002

Data Index: 741155 (C4 (Web Text))

Max Activating Token Index: 186

Text #5

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0791. Min Act: -0.0002

Data Index: 1176224 (C4 (Web Text))

Max Activating Token Index: 97

Text #6

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0758. Min Act: -0.0003

Data Index: 916307 (C4 (Web Text))

Max Activating Token Index: 5

Text #7

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0758. Min Act: -0.0003

Data Index: 676713 (C4 (Web Text))

Max Activating Token Index: 5

Text #8

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0776. Min Act: -0.0002

Data Index: 1084167 (C4 (Web Text))

Max Activating Token Index: 454

Text #9

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0770. Min Act: -0.0002

Data Index: 272551 (C4 (Web Text))

Max Activating Token Index: 24

Text #10

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0773. Min Act: -0.0002

Data Index: 543373 (C4 (Web Text))

Max Activating Token Index: 20

Text #11

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0710. Min Act: -0.0002

Data Index: 1499689 (Python Code)

Max Activating Token Index: 16

Text #12

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0685. Min Act: -0.0002

Data Index: 1008160 (C4 (Web Text))

Max Activating Token Index: 115

Text #13

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0765. Min Act: -0.0002

Data Index: 793626 (C4 (Web Text))

Max Activating Token Index: 7

Text #14

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0755. Min Act: -0.0002

Data Index: 650290 (C4 (Web Text))

Max Activating Token Index: 106

Text #15

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0760. Min Act: -0.0002

Data Index: 569106 (C4 (Web Text))

Max Activating Token Index: 31

Text #16

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0799. Min Act: -0.0002

Data Index: 1606033 (Python Code)

Max Activating Token Index: 608

Text #17

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0762. Min Act: -0.0002

Data Index: 645832 (C4 (Web Text))

Max Activating Token Index: 660

Text #18

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0719. Min Act: -0.0002

Data Index: 32185 (C4 (Web Text))

Max Activating Token Index: 60

Text #19

Max Range: 0.0891. Min Range: -0.0891

Max Act: 0.0689. Min Act: -0.0003

Data Index: 1156173 (C4 (Web Text))

Max Activating Token Index: 424