Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 459 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.3382. Min Range: -1.3382

Max Act: 1.3382. Min Act: -0.0000

Data Index: 1446818 (Python Code)

Max Activating Token Index: 843

Text #1

Max Range: 1.3382. Min Range: -1.3382

Max Act: 1.1724. Min Act: -0.0000

Data Index: 1624732 (Python Code)

Max Activating Token Index: 878

Text #2

Max Range: 1.3382. Min Range: -1.3382

Max Act: 1.0987. Min Act: -0.0000

Data Index: 1084966 (C4 (Web Text))

Max Activating Token Index: 192

Text #3

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.9609. Min Act: -0.0000

Data Index: 1142630 (C4 (Web Text))

Max Activating Token Index: 614

Text #4

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.9424. Min Act: -0.0000

Data Index: 1522310 (Python Code)

Max Activating Token Index: 10

Text #5

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.9751. Min Act: -0.0000

Data Index: 1637348 (Python Code)

Max Activating Token Index: 65

Text #6

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.9425. Min Act: -0.0000

Data Index: 1264535 (C4 (Web Text))

Max Activating Token Index: 633

Text #7

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.9588. Min Act: -0.0000

Data Index: 1494289 (Python Code)

Max Activating Token Index: 858

Text #8

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.9084. Min Act: -0.0000

Data Index: 281755 (C4 (Web Text))

Max Activating Token Index: 568

Text #9

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8756. Min Act: -0.0000

Data Index: 1612700 (Python Code)

Max Activating Token Index: 118

Text #10

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8366. Min Act: -0.0000

Data Index: 595739 (C4 (Web Text))

Max Activating Token Index: 61

Text #11

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8886. Min Act: -0.0000

Data Index: 1535109 (Python Code)

Max Activating Token Index: 164

Text #12

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8362. Min Act: -0.0000

Data Index: 1565718 (Python Code)

Max Activating Token Index: 20

Text #13

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8467. Min Act: -0.0000

Data Index: 1459941 (Python Code)

Max Activating Token Index: 1006

Text #14

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8128. Min Act: -0.0000

Data Index: 1472680 (Python Code)

Max Activating Token Index: 620

Text #15

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.7754. Min Act: -0.0000

Data Index: 1470461 (Python Code)

Max Activating Token Index: 109

Text #16

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8181. Min Act: -0.0000

Data Index: 1506001 (Python Code)

Max Activating Token Index: 566

Text #17

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.8147. Min Act: -0.0000

Data Index: 753757 (C4 (Web Text))

Max Activating Token Index: 524

Text #18

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.7971. Min Act: -0.0000

Data Index: 774963 (C4 (Web Text))

Max Activating Token Index: 75

Text #19

Max Range: 1.3382. Min Range: -1.3382

Max Act: 0.7698. Min Act: -0.0000

Data Index: 1649099 (Python Code)

Max Activating Token Index: 168