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

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

Neuron 3466 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0972. Min Act: -0.0001

Data Index: 306443 (C4 (Web Text))

Max Activating Token Index: 632

Text #1

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0888. Min Act: -0.0001

Data Index: 568325 (C4 (Web Text))

Max Activating Token Index: 635

Text #2

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0874. Min Act: -0.0001

Data Index: 1256229 (C4 (Web Text))

Max Activating Token Index: 856

Text #3

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0855. Min Act: -0.0001

Data Index: 737989 (C4 (Web Text))

Max Activating Token Index: 8

Text #4

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0731. Min Act: -0.0001

Data Index: 1035095 (C4 (Web Text))

Max Activating Token Index: 161

Text #5

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0728. Min Act: -0.0001

Data Index: 958393 (C4 (Web Text))

Max Activating Token Index: 149

Text #6

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0799. Min Act: -0.0001

Data Index: 575520 (C4 (Web Text))

Max Activating Token Index: 180

Text #7

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0842. Min Act: -0.0001

Data Index: 343669 (C4 (Web Text))

Max Activating Token Index: 357

Text #8

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0657. Min Act: -0.0001

Data Index: 240758 (C4 (Web Text))

Max Activating Token Index: 111

Text #9

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0792. Min Act: -0.0001

Data Index: 486064 (C4 (Web Text))

Max Activating Token Index: 212

Text #10

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0683. Min Act: -0.0001

Data Index: 899539 (C4 (Web Text))

Max Activating Token Index: 650

Text #11

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0599. Min Act: -0.0001

Data Index: 398192 (C4 (Web Text))

Max Activating Token Index: 123

Text #12

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0601. Min Act: -0.0001

Data Index: 489178 (C4 (Web Text))

Max Activating Token Index: 706

Text #13

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0580. Min Act: -0.0001

Data Index: 1501527 (Python Code)

Max Activating Token Index: 12

Text #14

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0616. Min Act: -0.0001

Data Index: 823309 (C4 (Web Text))

Max Activating Token Index: 24

Text #15

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0580. Min Act: -0.0001

Data Index: 587605 (C4 (Web Text))

Max Activating Token Index: 473

Text #16

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0589. Min Act: -0.0001

Data Index: 1355022 (C4 (Web Text))

Max Activating Token Index: 858

Text #17

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0567. Min Act: -0.0001

Data Index: 781650 (C4 (Web Text))

Max Activating Token Index: 629

Text #18

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0629. Min Act: -0.0001

Data Index: 294570 (C4 (Web Text))

Max Activating Token Index: 566

Text #19

Max Range: 0.0972. Min Range: -0.0972

Max Act: 0.0611. Min Act: -0.0001

Data Index: 1340452 (C4 (Web Text))

Max Activating Token Index: 740