Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1992 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0811. Min Act: -0.0005

Data Index: 1198808 (C4 (Web Text))

Max Activating Token Index: 981

Text #1

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0678. Min Act: -0.0005

Data Index: 881890 (C4 (Web Text))

Max Activating Token Index: 44

Text #2

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0653. Min Act: -0.0005

Data Index: 587333 (C4 (Web Text))

Max Activating Token Index: 916

Text #3

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0604. Min Act: -0.0005

Data Index: 693793 (C4 (Web Text))

Max Activating Token Index: 613

Text #4

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0622. Min Act: -0.0005

Data Index: 187261 (C4 (Web Text))

Max Activating Token Index: 586

Text #5

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0512. Min Act: -0.0005

Data Index: 1131060 (C4 (Web Text))

Max Activating Token Index: 667

Text #6

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0558. Min Act: -0.0005

Data Index: 1028656 (C4 (Web Text))

Max Activating Token Index: 742

Text #7

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0567. Min Act: -0.0005

Data Index: 267855 (C4 (Web Text))

Max Activating Token Index: 744

Text #8

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0547. Min Act: -0.0005

Data Index: 173485 (C4 (Web Text))

Max Activating Token Index: 891

Text #9

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0559. Min Act: -0.0005

Data Index: 318594 (C4 (Web Text))

Max Activating Token Index: 578

Text #10

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0531. Min Act: -0.0005

Data Index: 1305857 (C4 (Web Text))

Max Activating Token Index: 163

Text #11

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0532. Min Act: -0.0005

Data Index: 15750 (C4 (Web Text))

Max Activating Token Index: 475

Text #12

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0534. Min Act: -0.0005

Data Index: 830805 (C4 (Web Text))

Max Activating Token Index: 200

Text #13

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0493. Min Act: -0.0005

Data Index: 1651870 (Python Code)

Max Activating Token Index: 447

Text #14

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0505. Min Act: -0.0005

Data Index: 725105 (C4 (Web Text))

Max Activating Token Index: 108

Text #15

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0505. Min Act: -0.0005

Data Index: 198295 (C4 (Web Text))

Max Activating Token Index: 413

Text #16

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0526. Min Act: -0.0005

Data Index: 1292644 (C4 (Web Text))

Max Activating Token Index: 968

Text #17

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0519. Min Act: -0.0005

Data Index: 725162 (C4 (Web Text))

Max Activating Token Index: 532

Text #18

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0512. Min Act: -0.0005

Data Index: 653463 (C4 (Web Text))

Max Activating Token Index: 755

Text #19

Max Range: 0.0811. Min Range: -0.0811

Max Act: 0.0479. Min Act: -0.0005

Data Index: 1059691 (C4 (Web Text))

Max Activating Token Index: 612