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

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

Neuron 498 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.4639. Min Act: -0.0000

Data Index: 1483029 (Python Code)

Max Activating Token Index: 966

Text #1

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.4620. Min Act: -0.0000

Data Index: 1629477 (Python Code)

Max Activating Token Index: 334

Text #2

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3758. Min Act: -0.0000

Data Index: 1474812 (Python Code)

Max Activating Token Index: 628

Text #3

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3879. Min Act: -0.0000

Data Index: 1419336 (Python Code)

Max Activating Token Index: 571

Text #4

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3567. Min Act: -0.0000

Data Index: 1354281 (C4 (Web Text))

Max Activating Token Index: 952

Text #5

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3706. Min Act: -0.0000

Data Index: 1532986 (Python Code)

Max Activating Token Index: 598

Text #6

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3662. Min Act: -0.0000

Data Index: 714167 (C4 (Web Text))

Max Activating Token Index: 34

Text #7

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3486. Min Act: -0.0000

Data Index: 1232433 (C4 (Web Text))

Max Activating Token Index: 578

Text #8

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3228. Min Act: -0.0000

Data Index: 378531 (C4 (Web Text))

Max Activating Token Index: 924

Text #9

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3574. Min Act: -0.0000

Data Index: 618077 (C4 (Web Text))

Max Activating Token Index: 861

Text #10

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3299. Min Act: -0.0000

Data Index: 1397012 (Python Code)

Max Activating Token Index: 221

Text #11

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3442. Min Act: -0.0000

Data Index: 1208450 (C4 (Web Text))

Max Activating Token Index: 915

Text #12

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3255. Min Act: -0.0000

Data Index: 1584251 (Python Code)

Max Activating Token Index: 453

Text #13

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3521. Min Act: -0.0000

Data Index: 1428823 (Python Code)

Max Activating Token Index: 136

Text #14

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3182. Min Act: -0.0000

Data Index: 1476346 (Python Code)

Max Activating Token Index: 276

Text #15

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.2719. Min Act: -0.0000

Data Index: 1503192 (Python Code)

Max Activating Token Index: 682

Text #16

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3159. Min Act: -0.0000

Data Index: 1561371 (Python Code)

Max Activating Token Index: 586

Text #17

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3124. Min Act: -0.0000

Data Index: 722960 (C4 (Web Text))

Max Activating Token Index: 715

Text #18

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3144. Min Act: -0.0000

Data Index: 982410 (C4 (Web Text))

Max Activating Token Index: 458

Text #19

Max Range: 0.4639. Min Range: -0.4639

Max Act: 0.3017. Min Act: -0.0000

Data Index: 1472784 (Python Code)

Max Activating Token Index: 222