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

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

Neuron 676 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: 0.0565. Min Range: -0.0565

Max Act: 0.0565. Min Act: -0.0001

Data Index: 1519291 (Python Code)

Max Activating Token Index: 978

Text #1

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0532. Min Act: -0.0001

Data Index: 1595506 (Python Code)

Max Activating Token Index: 953

Text #2

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0526. Min Act: -0.0003

Data Index: 84611 (C4 (Web Text))

Max Activating Token Index: 73

Text #3

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0497. Min Act: -0.0003

Data Index: 600406 (C4 (Web Text))

Max Activating Token Index: 350

Text #4

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0481. Min Act: -0.0002

Data Index: 1396636 (Python Code)

Max Activating Token Index: 950

Text #5

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0423. Min Act: -0.0002

Data Index: 1404496 (Python Code)

Max Activating Token Index: 563

Text #6

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0420. Min Act: -0.0003

Data Index: 34738 (C4 (Web Text))

Max Activating Token Index: 859

Text #7

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0433. Min Act: -0.0003

Data Index: 384643 (C4 (Web Text))

Max Activating Token Index: 931

Text #8

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0402. Min Act: -0.0003

Data Index: 1579015 (Python Code)

Max Activating Token Index: 947

Text #9

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0433. Min Act: -0.0002

Data Index: 290792 (C4 (Web Text))

Max Activating Token Index: 187

Text #10

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0432. Min Act: -0.0003

Data Index: 1088644 (C4 (Web Text))

Max Activating Token Index: 826

Text #11

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0412. Min Act: -0.0002

Data Index: 1376177 (Python Code)

Max Activating Token Index: 998

Text #12

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0419. Min Act: -0.0002

Data Index: 1654056 (Python Code)

Max Activating Token Index: 890

Text #13

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0426. Min Act: -0.0003

Data Index: 713748 (C4 (Web Text))

Max Activating Token Index: 958

Text #14

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0407. Min Act: -0.0003

Data Index: 1277493 (C4 (Web Text))

Max Activating Token Index: 813

Text #15

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0413. Min Act: -0.0002

Data Index: 1499044 (Python Code)

Max Activating Token Index: 165

Text #16

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0408. Min Act: -0.0003

Data Index: 1250765 (C4 (Web Text))

Max Activating Token Index: 750

Text #17

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0416. Min Act: -0.0003

Data Index: 1534087 (Python Code)

Max Activating Token Index: 771

Text #18

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0408. Min Act: -0.0003

Data Index: 1091160 (C4 (Web Text))

Max Activating Token Index: 571

Text #19

Max Range: 0.0565. Min Range: -0.0565

Max Act: 0.0412. Min Act: -0.0002

Data Index: 1365962 (Python Code)

Max Activating Token Index: 766