Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 33 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0859. Min Act: -0.0003

Data Index: 22222 (C4 (Web Text))

Max Activating Token Index: 893

Text #1

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0714. Min Act: -0.0003

Data Index: 948358 (C4 (Web Text))

Max Activating Token Index: 692

Text #2

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0815. Min Act: -0.0003

Data Index: 830464 (C4 (Web Text))

Max Activating Token Index: 222

Text #3

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0695. Min Act: -0.0003

Data Index: 784454 (C4 (Web Text))

Max Activating Token Index: 417

Text #4

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0676. Min Act: -0.0003

Data Index: 983321 (C4 (Web Text))

Max Activating Token Index: 198

Text #5

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0676. Min Act: -0.0003

Data Index: 461655 (C4 (Web Text))

Max Activating Token Index: 831

Text #6

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0670. Min Act: -0.0003

Data Index: 260584 (C4 (Web Text))

Max Activating Token Index: 644

Text #7

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0632. Min Act: -0.0003

Data Index: 1218418 (C4 (Web Text))

Max Activating Token Index: 689

Text #8

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0583. Min Act: -0.0003

Data Index: 240310 (C4 (Web Text))

Max Activating Token Index: 948

Text #9

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0572. Min Act: -0.0003

Data Index: 362023 (C4 (Web Text))

Max Activating Token Index: 820

Text #10

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0577. Min Act: -0.0003

Data Index: 643192 (C4 (Web Text))

Max Activating Token Index: 976

Text #11

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0558. Min Act: -0.0003

Data Index: 692393 (C4 (Web Text))

Max Activating Token Index: 81

Text #12

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0559. Min Act: -0.0003

Data Index: 606286 (C4 (Web Text))

Max Activating Token Index: 240

Text #13

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0548. Min Act: -0.0003

Data Index: 618382 (C4 (Web Text))

Max Activating Token Index: 500

Text #14

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0593. Min Act: -0.0003

Data Index: 330382 (C4 (Web Text))

Max Activating Token Index: 391

Text #15

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0541. Min Act: -0.0003

Data Index: 1249341 (C4 (Web Text))

Max Activating Token Index: 518

Text #16

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0516. Min Act: -0.0003

Data Index: 922094 (C4 (Web Text))

Max Activating Token Index: 572

Text #17

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0525. Min Act: -0.0003

Data Index: 905097 (C4 (Web Text))

Max Activating Token Index: 913

Text #18

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0517. Min Act: -0.0003

Data Index: 1228922 (C4 (Web Text))

Max Activating Token Index: 585

Text #19

Max Range: 0.0859. Min Range: -0.0859

Max Act: 0.0521. Min Act: -0.0003

Data Index: 1349822 (C4 (Web Text))

Max Activating Token Index: 964