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

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

Neuron 4822 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0656. Min Act: -0.0003

Data Index: 1093645 (C4 (Web Text))

Max Activating Token Index: 119

Text #1

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0602. Min Act: -0.0003

Data Index: 586 (C4 (Web Text))

Max Activating Token Index: 744

Text #2

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0601. Min Act: -0.0003

Data Index: 739771 (C4 (Web Text))

Max Activating Token Index: 868

Text #3

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0589. Min Act: -0.0003

Data Index: 393663 (C4 (Web Text))

Max Activating Token Index: 931

Text #4

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0581. Min Act: -0.0003

Data Index: 130833 (C4 (Web Text))

Max Activating Token Index: 484

Text #5

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0585. Min Act: -0.0002

Data Index: 896554 (C4 (Web Text))

Max Activating Token Index: 514

Text #6

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0582. Min Act: -0.0003

Data Index: 585663 (C4 (Web Text))

Max Activating Token Index: 119

Text #7

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0593. Min Act: -0.0002

Data Index: 1413514 (Python Code)

Max Activating Token Index: 65

Text #8

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0586. Min Act: -0.0002

Data Index: 1420021 (Python Code)

Max Activating Token Index: 564

Text #9

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0547. Min Act: -0.0003

Data Index: 1358512 (C4 (Web Text))

Max Activating Token Index: 333

Text #10

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0569. Min Act: -0.0003

Data Index: 359100 (C4 (Web Text))

Max Activating Token Index: 448

Text #11

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0548. Min Act: -0.0003

Data Index: 244366 (C4 (Web Text))

Max Activating Token Index: 464

Text #12

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0539. Min Act: -0.0003

Data Index: 8598 (C4 (Web Text))

Max Activating Token Index: 268

Text #13

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0556. Min Act: -0.0002

Data Index: 1493305 (Python Code)

Max Activating Token Index: 865

Text #14

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0516. Min Act: -0.0003

Data Index: 146880 (C4 (Web Text))

Max Activating Token Index: 571

Text #15

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0552. Min Act: -0.0002

Data Index: 508865 (C4 (Web Text))

Max Activating Token Index: 328

Text #16

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0602. Min Act: -0.0003

Data Index: 1000472 (C4 (Web Text))

Max Activating Token Index: 651

Text #17

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0587. Min Act: -0.0003

Data Index: 543854 (C4 (Web Text))

Max Activating Token Index: 120

Text #18

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0512. Min Act: -0.0003

Data Index: 740781 (C4 (Web Text))

Max Activating Token Index: 931

Text #19

Max Range: 0.0656. Min Range: -0.0656

Max Act: 0.0547. Min Act: -0.0003

Data Index: 267216 (C4 (Web Text))

Max Activating Token Index: 488