Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2023 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0733. Min Act: -0.0001

Data Index: 376173 (C4 (Web Text))

Max Activating Token Index: 674

Text #1

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0704. Min Act: -0.0001

Data Index: 212274 (C4 (Web Text))

Max Activating Token Index: 251

Text #2

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0707. Min Act: -0.0001

Data Index: 940794 (C4 (Web Text))

Max Activating Token Index: 386

Text #3

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0679. Min Act: -0.0001

Data Index: 1572028 (Python Code)

Max Activating Token Index: 539

Text #4

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0565. Min Act: -0.0001

Data Index: 930859 (C4 (Web Text))

Max Activating Token Index: 669

Text #5

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0620. Min Act: -0.0001

Data Index: 404518 (C4 (Web Text))

Max Activating Token Index: 748

Text #6

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0591. Min Act: -0.0000

Data Index: 1008151 (C4 (Web Text))

Max Activating Token Index: 719

Text #7

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0607. Min Act: -0.0000

Data Index: 120061 (C4 (Web Text))

Max Activating Token Index: 493

Text #8

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0602. Min Act: -0.0001

Data Index: 625101 (C4 (Web Text))

Max Activating Token Index: 385

Text #9

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0575. Min Act: -0.0000

Data Index: 409870 (C4 (Web Text))

Max Activating Token Index: 71

Text #10

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0529. Min Act: -0.0001

Data Index: 1452594 (Python Code)

Max Activating Token Index: 1010

Text #11

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0618. Min Act: -0.0001

Data Index: 1569390 (Python Code)

Max Activating Token Index: 395

Text #12

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0542. Min Act: -0.0000

Data Index: 667994 (C4 (Web Text))

Max Activating Token Index: 69

Text #13

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0542. Min Act: -0.0001

Data Index: 1270915 (C4 (Web Text))

Max Activating Token Index: 840

Text #14

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0588. Min Act: -0.0001

Data Index: 855404 (C4 (Web Text))

Max Activating Token Index: 899

Text #15

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0594. Min Act: -0.0001

Data Index: 665630 (C4 (Web Text))

Max Activating Token Index: 643

Text #16

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0580. Min Act: -0.0001

Data Index: 357615 (C4 (Web Text))

Max Activating Token Index: 796

Text #17

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0553. Min Act: -0.0000

Data Index: 1546857 (Python Code)

Max Activating Token Index: 813

Text #18

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0603. Min Act: -0.0001

Data Index: 1458238 (Python Code)

Max Activating Token Index: 482

Text #19

Max Range: 0.0733. Min Range: -0.0733

Max Act: 0.0567. Min Act: -0.0001

Data Index: 1249130 (C4 (Web Text))

Max Activating Token Index: 814