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

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

Neuron 54 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.3709. Min Act: -0.0000

Data Index: 747129 (C4 (Web Text))

Max Activating Token Index: 653

Text #1

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.3368. Min Act: -0.0000

Data Index: 509683 (C4 (Web Text))

Max Activating Token Index: 586

Text #2

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.3029. Min Act: -0.0000

Data Index: 606965 (C4 (Web Text))

Max Activating Token Index: 194

Text #3

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2867. Min Act: -0.0000

Data Index: 1571507 (Python Code)

Max Activating Token Index: 573

Text #4

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2817. Min Act: -0.0000

Data Index: 1365671 (Python Code)

Max Activating Token Index: 759

Text #5

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2629. Min Act: -0.0000

Data Index: 1473123 (Python Code)

Max Activating Token Index: 615

Text #6

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2766. Min Act: -0.0000

Data Index: 1480136 (Python Code)

Max Activating Token Index: 496

Text #7

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2707. Min Act: -0.0000

Data Index: 1243282 (C4 (Web Text))

Max Activating Token Index: 182

Text #8

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2931. Min Act: -0.0000

Data Index: 744418 (C4 (Web Text))

Max Activating Token Index: 554

Text #9

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2638. Min Act: -0.0000

Data Index: 1313402 (C4 (Web Text))

Max Activating Token Index: 322

Text #10

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2585. Min Act: -0.0000

Data Index: 54582 (C4 (Web Text))

Max Activating Token Index: 808

Text #11

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2369. Min Act: -0.0000

Data Index: 1541827 (Python Code)

Max Activating Token Index: 126

Text #12

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2406. Min Act: -0.0000

Data Index: 604784 (C4 (Web Text))

Max Activating Token Index: 585

Text #13

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2304. Min Act: -0.0000

Data Index: 1578893 (Python Code)

Max Activating Token Index: 55

Text #14

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2410. Min Act: -0.0000

Data Index: 1019981 (C4 (Web Text))

Max Activating Token Index: 26

Text #15

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2518. Min Act: -0.0000

Data Index: 1274140 (C4 (Web Text))

Max Activating Token Index: 968

Text #16

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2302. Min Act: -0.0000

Data Index: 1484233 (Python Code)

Max Activating Token Index: 547

Text #17

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2259. Min Act: -0.0000

Data Index: 520110 (C4 (Web Text))

Max Activating Token Index: 530

Text #18

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2219. Min Act: -0.0000

Data Index: 1390824 (Python Code)

Max Activating Token Index: 617

Text #19

Max Range: 0.3709. Min Range: -0.3709

Max Act: 0.2211. Min Act: -0.0000

Data Index: 383421 (C4 (Web Text))

Max Activating Token Index: 342