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

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

Neuron 126 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: 0.0633. Min Range: -0.0633

Max Act: 0.0633. Min Act: -0.0003

Data Index: 1087822 (C4 (Web Text))

Max Activating Token Index: 80

Text #1

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0581. Min Act: -0.0003

Data Index: 206102 (C4 (Web Text))

Max Activating Token Index: 428

Text #2

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0543. Min Act: -0.0003

Data Index: 666660 (C4 (Web Text))

Max Activating Token Index: 85

Text #3

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0553. Min Act: -0.0003

Data Index: 216879 (C4 (Web Text))

Max Activating Token Index: 436

Text #4

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0532. Min Act: -0.0004

Data Index: 835326 (C4 (Web Text))

Max Activating Token Index: 130

Text #5

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0537. Min Act: -0.0004

Data Index: 769554 (C4 (Web Text))

Max Activating Token Index: 147

Text #6

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0534. Min Act: -0.0003

Data Index: 92714 (C4 (Web Text))

Max Activating Token Index: 371

Text #7

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0532. Min Act: -0.0003

Data Index: 124393 (C4 (Web Text))

Max Activating Token Index: 264

Text #8

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0516. Min Act: -0.0002

Data Index: 701135 (C4 (Web Text))

Max Activating Token Index: 126

Text #9

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0496. Min Act: -0.0004

Data Index: 346967 (C4 (Web Text))

Max Activating Token Index: 574

Text #10

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0509. Min Act: -0.0002

Data Index: 623767 (C4 (Web Text))

Max Activating Token Index: 932

Text #11

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0483. Min Act: -0.0003

Data Index: 1186139 (C4 (Web Text))

Max Activating Token Index: 704

Text #12

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0487. Min Act: -0.0004

Data Index: 376890 (C4 (Web Text))

Max Activating Token Index: 600

Text #13

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0490. Min Act: -0.0003

Data Index: 1026997 (C4 (Web Text))

Max Activating Token Index: 71

Text #14

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0488. Min Act: -0.0003

Data Index: 51439 (C4 (Web Text))

Max Activating Token Index: 637

Text #15

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0465. Min Act: -0.0002

Data Index: 1335095 (C4 (Web Text))

Max Activating Token Index: 655

Text #16

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0473. Min Act: -0.0003

Data Index: 837975 (C4 (Web Text))

Max Activating Token Index: 179

Text #17

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0464. Min Act: -0.0003

Data Index: 1087010 (C4 (Web Text))

Max Activating Token Index: 562

Text #18

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0463. Min Act: -0.0004

Data Index: 701400 (C4 (Web Text))

Max Activating Token Index: 671

Text #19

Max Range: 0.0633. Min Range: -0.0633

Max Act: 0.0485. Min Act: -0.0003

Data Index: 280894 (C4 (Web Text))

Max Activating Token Index: 228