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

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

Neuron 3314 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: 4.5730. Min Range: -4.5730

Max Act: 4.5730. Min Act: -0.0000

Data Index: 1075665 (C4 (Web Text))

Max Activating Token Index: 331

Text #1

Max Range: 4.5730. Min Range: -4.5730

Max Act: 4.4926. Min Act: 0.0000

Data Index: 936002 (C4 (Web Text))

Max Activating Token Index: 212

Text #2

Max Range: 4.5730. Min Range: -4.5730

Max Act: 3.9269. Min Act: -0.0000

Data Index: 608024 (C4 (Web Text))

Max Activating Token Index: 359

Text #3

Max Range: 4.5730. Min Range: -4.5730

Max Act: 3.1852. Min Act: -0.0000

Data Index: 1604395 (Python Code)

Max Activating Token Index: 812

Text #4

Max Range: 4.5730. Min Range: -4.5730

Max Act: 3.3197. Min Act: -0.0000

Data Index: 47078 (C4 (Web Text))

Max Activating Token Index: 403

Text #5

Max Range: 4.5730. Min Range: -4.5730

Max Act: 3.0227. Min Act: -0.0000

Data Index: 176008 (C4 (Web Text))

Max Activating Token Index: 849

Text #6

Max Range: 4.5730. Min Range: -4.5730

Max Act: 3.1082. Min Act: -0.0000

Data Index: 1264173 (C4 (Web Text))

Max Activating Token Index: 918

Text #7

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.6556. Min Act: -0.0000

Data Index: 1427880 (Python Code)

Max Activating Token Index: 441

Text #8

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.5679. Min Act: 0.0000

Data Index: 598967 (C4 (Web Text))

Max Activating Token Index: 893

Text #9

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.3582. Min Act: -0.0000

Data Index: 572535 (C4 (Web Text))

Max Activating Token Index: 78

Text #10

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.2822. Min Act: -0.0000

Data Index: 312945 (C4 (Web Text))

Max Activating Token Index: 496

Text #11

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.2772. Min Act: -0.0000

Data Index: 372216 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.2462. Min Act: -0.0000

Data Index: 1232020 (C4 (Web Text))

Max Activating Token Index: 248

Text #13

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.2412. Min Act: -0.0000

Data Index: 1115917 (C4 (Web Text))

Max Activating Token Index: 349

Text #14

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.3104. Min Act: -0.0000

Data Index: 486376 (C4 (Web Text))

Max Activating Token Index: 563

Text #15

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.3933. Min Act: -0.0001

Data Index: 1459508 (Python Code)

Max Activating Token Index: 679

Text #16

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.0900. Min Act: 0.0000

Data Index: 601764 (C4 (Web Text))

Max Activating Token Index: 571

Text #17

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.0981. Min Act: -0.0000

Data Index: 408903 (C4 (Web Text))

Max Activating Token Index: 924

Text #18

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.2743. Min Act: -0.0000

Data Index: 1566653 (Python Code)

Max Activating Token Index: 506

Text #19

Max Range: 4.5730. Min Range: -4.5730

Max Act: 2.2101. Min Act: -0.0000

Data Index: 1107792 (C4 (Web Text))

Max Activating Token Index: 251