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

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

Neuron 3940 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: 0.0726. Min Range: -0.0726

Max Act: 0.0726. Min Act: -0.0002

Data Index: 1025828 (C4 (Web Text))

Max Activating Token Index: 426

Text #1

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0529. Min Act: -0.0002

Data Index: 87402 (C4 (Web Text))

Max Activating Token Index: 392

Text #2

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0526. Min Act: -0.0002

Data Index: 335142 (C4 (Web Text))

Max Activating Token Index: 716

Text #3

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0541. Min Act: -0.0002

Data Index: 126708 (C4 (Web Text))

Max Activating Token Index: 707

Text #4

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0513. Min Act: -0.0002

Data Index: 321 (C4 (Web Text))

Max Activating Token Index: 396

Text #5

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0499. Min Act: -0.0002

Data Index: 923499 (C4 (Web Text))

Max Activating Token Index: 865

Text #6

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0521. Min Act: -0.0002

Data Index: 77360 (C4 (Web Text))

Max Activating Token Index: 1010

Text #7

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0502. Min Act: -0.0002

Data Index: 444612 (C4 (Web Text))

Max Activating Token Index: 898

Text #8

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0508. Min Act: -0.0002

Data Index: 1308997 (C4 (Web Text))

Max Activating Token Index: 556

Text #9

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0486. Min Act: -0.0002

Data Index: 42617 (C4 (Web Text))

Max Activating Token Index: 839

Text #10

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0463. Min Act: -0.0002

Data Index: 441626 (C4 (Web Text))

Max Activating Token Index: 528

Text #11

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0450. Min Act: -0.0002

Data Index: 1478295 (Python Code)

Max Activating Token Index: 681

Text #12

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0464. Min Act: -0.0002

Data Index: 623228 (C4 (Web Text))

Max Activating Token Index: 723

Text #13

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0464. Min Act: -0.0002

Data Index: 1065881 (C4 (Web Text))

Max Activating Token Index: 887

Text #14

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0467. Min Act: -0.0002

Data Index: 1037814 (C4 (Web Text))

Max Activating Token Index: 674

Text #15

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0441. Min Act: -0.0002

Data Index: 1396343 (Python Code)

Max Activating Token Index: 301

Text #16

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0447. Min Act: -0.0002

Data Index: 134923 (C4 (Web Text))

Max Activating Token Index: 704

Text #17

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0431. Min Act: -0.0002

Data Index: 1551665 (Python Code)

Max Activating Token Index: 205

Text #18

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0438. Min Act: -0.0002

Data Index: 1192033 (C4 (Web Text))

Max Activating Token Index: 20

Text #19

Max Range: 0.0726. Min Range: -0.0726

Max Act: 0.0424. Min Act: -0.0002

Data Index: 165586 (C4 (Web Text))

Max Activating Token Index: 804