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

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

Neuron 5076 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: 0.1056. Min Range: -0.1056

Max Act: 0.1056. Min Act: -0.0000

Data Index: 585822 (C4 (Web Text))

Max Activating Token Index: 129

Text #1

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.1048. Min Act: -0.0000

Data Index: 1107901 (C4 (Web Text))

Max Activating Token Index: 792

Text #2

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0965. Min Act: -0.0000

Data Index: 486038 (C4 (Web Text))

Max Activating Token Index: 751

Text #3

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0928. Min Act: -0.0000

Data Index: 826122 (C4 (Web Text))

Max Activating Token Index: 551

Text #4

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0911. Min Act: -0.0000

Data Index: 583038 (C4 (Web Text))

Max Activating Token Index: 823

Text #5

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0946. Min Act: -0.0000

Data Index: 215788 (C4 (Web Text))

Max Activating Token Index: 778

Text #6

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0909. Min Act: -0.0000

Data Index: 610314 (C4 (Web Text))

Max Activating Token Index: 328

Text #7

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0906. Min Act: -0.0000

Data Index: 1264239 (C4 (Web Text))

Max Activating Token Index: 101

Text #8

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0840. Min Act: -0.0000

Data Index: 804551 (C4 (Web Text))

Max Activating Token Index: 244

Text #9

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0871. Min Act: -0.0000

Data Index: 213785 (C4 (Web Text))

Max Activating Token Index: 200

Text #10

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0896. Min Act: -0.0000

Data Index: 778841 (C4 (Web Text))

Max Activating Token Index: 346

Text #11

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0822. Min Act: -0.0000

Data Index: 24083 (C4 (Web Text))

Max Activating Token Index: 398

Text #12

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0866. Min Act: -0.0000

Data Index: 577595 (C4 (Web Text))

Max Activating Token Index: 901

Text #13

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0786. Min Act: -0.0000

Data Index: 1074634 (C4 (Web Text))

Max Activating Token Index: 327

Text #14

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0778. Min Act: -0.0000

Data Index: 1383122 (Python Code)

Max Activating Token Index: 445

Text #15

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0805. Min Act: -0.0000

Data Index: 1126116 (C4 (Web Text))

Max Activating Token Index: 76

Text #16

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0831. Min Act: -0.0000

Data Index: 897401 (C4 (Web Text))

Max Activating Token Index: 972

Text #17

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0815. Min Act: -0.0000

Data Index: 389320 (C4 (Web Text))

Max Activating Token Index: 75

Text #18

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0758. Min Act: -0.0000

Data Index: 925652 (C4 (Web Text))

Max Activating Token Index: 54

Text #19

Max Range: 0.1056. Min Range: -0.1056

Max Act: 0.0761. Min Act: -0.0000

Data Index: 1115120 (C4 (Web Text))

Max Activating Token Index: 156