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

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

Neuron 198 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: 0.0735. Min Range: -0.0735

Max Act: 0.0735. Min Act: -0.0003

Data Index: 149617 (C4 (Web Text))

Max Activating Token Index: 1001

Text #1

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0635. Min Act: -0.0003

Data Index: 918596 (C4 (Web Text))

Max Activating Token Index: 39

Text #2

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0620. Min Act: -0.0003

Data Index: 740466 (C4 (Web Text))

Max Activating Token Index: 163

Text #3

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0593. Min Act: -0.0003

Data Index: 995428 (C4 (Web Text))

Max Activating Token Index: 629

Text #4

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0614. Min Act: -0.0004

Data Index: 406101 (C4 (Web Text))

Max Activating Token Index: 401

Text #5

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0594. Min Act: -0.0002

Data Index: 924028 (C4 (Web Text))

Max Activating Token Index: 476

Text #6

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0602. Min Act: -0.0003

Data Index: 103275 (C4 (Web Text))

Max Activating Token Index: 735

Text #7

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0583. Min Act: -0.0003

Data Index: 131876 (C4 (Web Text))

Max Activating Token Index: 203

Text #8

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0596. Min Act: -0.0003

Data Index: 1427346 (Python Code)

Max Activating Token Index: 694

Text #9

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0587. Min Act: -0.0003

Data Index: 662690 (C4 (Web Text))

Max Activating Token Index: 420

Text #10

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0566. Min Act: -0.0003

Data Index: 313546 (C4 (Web Text))

Max Activating Token Index: 1003

Text #11

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0563. Min Act: -0.0003

Data Index: 1131067 (C4 (Web Text))

Max Activating Token Index: 455

Text #12

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0550. Min Act: -0.0003

Data Index: 998987 (C4 (Web Text))

Max Activating Token Index: 206

Text #13

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0560. Min Act: -0.0003

Data Index: 819908 (C4 (Web Text))

Max Activating Token Index: 248

Text #14

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0574. Min Act: -0.0003

Data Index: 701326 (C4 (Web Text))

Max Activating Token Index: 808

Text #15

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0544. Min Act: -0.0003

Data Index: 248662 (C4 (Web Text))

Max Activating Token Index: 131

Text #16

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0529. Min Act: -0.0003

Data Index: 586251 (C4 (Web Text))

Max Activating Token Index: 797

Text #17

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0546. Min Act: -0.0002

Data Index: 447990 (C4 (Web Text))

Max Activating Token Index: 418

Text #18

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0563. Min Act: -0.0003

Data Index: 688929 (C4 (Web Text))

Max Activating Token Index: 958

Text #19

Max Range: 0.0735. Min Range: -0.0735

Max Act: 0.0533. Min Act: -0.0002

Data Index: 392824 (C4 (Web Text))

Max Activating Token Index: 355