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

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

Neuron 4215 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.2632. Min Act: -0.0000

Data Index: 125193 (C4 (Web Text))

Max Activating Token Index: 160

Text #1

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.2290. Min Act: -0.0000

Data Index: 443466 (C4 (Web Text))

Max Activating Token Index: 544

Text #2

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1948. Min Act: -0.0000

Data Index: 369433 (C4 (Web Text))

Max Activating Token Index: 868

Text #3

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.2028. Min Act: -0.0000

Data Index: 129434 (C4 (Web Text))

Max Activating Token Index: 274

Text #4

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.2080. Min Act: -0.0000

Data Index: 1407695 (Python Code)

Max Activating Token Index: 440

Text #5

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1993. Min Act: -0.0000

Data Index: 513694 (C4 (Web Text))

Max Activating Token Index: 510

Text #6

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.2006. Min Act: -0.0000

Data Index: 237032 (C4 (Web Text))

Max Activating Token Index: 886

Text #7

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.2023. Min Act: -0.0000

Data Index: 19469 (C4 (Web Text))

Max Activating Token Index: 395

Text #8

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1936. Min Act: -0.0000

Data Index: 1042962 (C4 (Web Text))

Max Activating Token Index: 473

Text #9

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1911. Min Act: -0.0000

Data Index: 625223 (C4 (Web Text))

Max Activating Token Index: 301

Text #10

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1869. Min Act: -0.0000

Data Index: 1336551 (C4 (Web Text))

Max Activating Token Index: 915

Text #11

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1775. Min Act: -0.0000

Data Index: 1232349 (C4 (Web Text))

Max Activating Token Index: 143

Text #12

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1879. Min Act: -0.0000

Data Index: 614806 (C4 (Web Text))

Max Activating Token Index: 487

Text #13

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1799. Min Act: -0.0000

Data Index: 431895 (C4 (Web Text))

Max Activating Token Index: 507

Text #14

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1851. Min Act: -0.0000

Data Index: 682801 (C4 (Web Text))

Max Activating Token Index: 265

Text #15

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1763. Min Act: -0.0000

Data Index: 837712 (C4 (Web Text))

Max Activating Token Index: 636

Text #16

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1755. Min Act: -0.0000

Data Index: 185104 (C4 (Web Text))

Max Activating Token Index: 29

Text #17

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1813. Min Act: -0.0000

Data Index: 710457 (C4 (Web Text))

Max Activating Token Index: 490

Text #18

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1732. Min Act: -0.0000

Data Index: 667356 (C4 (Web Text))

Max Activating Token Index: 77

Text #19

Max Range: 0.2632. Min Range: -0.2632

Max Act: 0.1692. Min Act: -0.0000

Data Index: 635342 (C4 (Web Text))

Max Activating Token Index: 16