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

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

Neuron 633 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: 0.0814. Min Range: -0.0814

Max Act: 0.0814. Min Act: -0.0003

Data Index: 856254 (C4 (Web Text))

Max Activating Token Index: 989

Text #1

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0745. Min Act: -0.0003

Data Index: 341652 (C4 (Web Text))

Max Activating Token Index: 876

Text #2

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0728. Min Act: -0.0003

Data Index: 9641 (C4 (Web Text))

Max Activating Token Index: 466

Text #3

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0729. Min Act: -0.0003

Data Index: 268264 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0730. Min Act: -0.0003

Data Index: 440867 (C4 (Web Text))

Max Activating Token Index: 272

Text #5

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0708. Min Act: -0.0003

Data Index: 367012 (C4 (Web Text))

Max Activating Token Index: 320

Text #6

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0706. Min Act: -0.0004

Data Index: 995447 (C4 (Web Text))

Max Activating Token Index: 434

Text #7

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0686. Min Act: -0.0003

Data Index: 246373 (C4 (Web Text))

Max Activating Token Index: 100

Text #8

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0687. Min Act: -0.0003

Data Index: 161909 (C4 (Web Text))

Max Activating Token Index: 651

Text #9

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0664. Min Act: -0.0003

Data Index: 945097 (C4 (Web Text))

Max Activating Token Index: 198

Text #10

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0680. Min Act: -0.0003

Data Index: 332364 (C4 (Web Text))

Max Activating Token Index: 240

Text #11

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0663. Min Act: -0.0003

Data Index: 675853 (C4 (Web Text))

Max Activating Token Index: 112

Text #12

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0676. Min Act: -0.0003

Data Index: 5011 (C4 (Web Text))

Max Activating Token Index: 593

Text #13

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0676. Min Act: -0.0003

Data Index: 810006 (C4 (Web Text))

Max Activating Token Index: 617

Text #14

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0653. Min Act: -0.0003

Data Index: 552679 (C4 (Web Text))

Max Activating Token Index: 211

Text #15

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0634. Min Act: -0.0003

Data Index: 1030173 (C4 (Web Text))

Max Activating Token Index: 817

Text #16

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0623. Min Act: -0.0003

Data Index: 292901 (C4 (Web Text))

Max Activating Token Index: 88

Text #17

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0643. Min Act: -0.0002

Data Index: 583096 (C4 (Web Text))

Max Activating Token Index: 315

Text #18

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0645. Min Act: -0.0003

Data Index: 311524 (C4 (Web Text))

Max Activating Token Index: 82

Text #19

Max Range: 0.0814. Min Range: -0.0814

Max Act: 0.0635. Min Act: -0.0004

Data Index: 4018 (C4 (Web Text))

Max Activating Token Index: 634