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

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

Neuron 4133 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: 0.0906. Min Range: -0.0906

Max Act: 0.0906. Min Act: -0.0002

Data Index: 18933 (C4 (Web Text))

Max Activating Token Index: 256

Text #1

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0888. Min Act: -0.0002

Data Index: 823082 (C4 (Web Text))

Max Activating Token Index: 897

Text #2

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0851. Min Act: -0.0002

Data Index: 703377 (C4 (Web Text))

Max Activating Token Index: 434

Text #3

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0746. Min Act: -0.0002

Data Index: 38776 (C4 (Web Text))

Max Activating Token Index: 88

Text #4

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0777. Min Act: -0.0002

Data Index: 388914 (C4 (Web Text))

Max Activating Token Index: 384

Text #5

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0731. Min Act: -0.0002

Data Index: 949398 (C4 (Web Text))

Max Activating Token Index: 288

Text #6

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0768. Min Act: -0.0003

Data Index: 225864 (C4 (Web Text))

Max Activating Token Index: 573

Text #7

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0748. Min Act: -0.0002

Data Index: 227190 (C4 (Web Text))

Max Activating Token Index: 520

Text #8

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0744. Min Act: -0.0002

Data Index: 1078727 (C4 (Web Text))

Max Activating Token Index: 564

Text #9

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0713. Min Act: -0.0002

Data Index: 64609 (C4 (Web Text))

Max Activating Token Index: 729

Text #10

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0683. Min Act: -0.0002

Data Index: 283078 (C4 (Web Text))

Max Activating Token Index: 752

Text #11

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0664. Min Act: -0.0002

Data Index: 1387077 (Python Code)

Max Activating Token Index: 31

Text #12

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0701. Min Act: -0.0002

Data Index: 906949 (C4 (Web Text))

Max Activating Token Index: 892

Text #13

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0681. Min Act: -0.0002

Data Index: 1493395 (Python Code)

Max Activating Token Index: 918

Text #14

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0673. Min Act: -0.0002

Data Index: 1296888 (C4 (Web Text))

Max Activating Token Index: 683

Text #15

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0691. Min Act: -0.0002

Data Index: 17007 (C4 (Web Text))

Max Activating Token Index: 217

Text #16

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0676. Min Act: -0.0002

Data Index: 229470 (C4 (Web Text))

Max Activating Token Index: 45

Text #17

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0647. Min Act: -0.0002

Data Index: 1645550 (Python Code)

Max Activating Token Index: 902

Text #18

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0674. Min Act: -0.0002

Data Index: 993796 (C4 (Web Text))

Max Activating Token Index: 305

Text #19

Max Range: 0.0906. Min Range: -0.0906

Max Act: 0.0662. Min Act: -0.0002

Data Index: 952422 (C4 (Web Text))

Max Activating Token Index: 5