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

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

Neuron 5500 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2604. Min Act: -0.0001

Data Index: 683328 (C4 (Web Text))

Max Activating Token Index: 675

Text #1

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2384. Min Act: -0.0001

Data Index: 167158 (C4 (Web Text))

Max Activating Token Index: 694

Text #2

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2267. Min Act: -0.0001

Data Index: 51218 (C4 (Web Text))

Max Activating Token Index: 1010

Text #3

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2306. Min Act: -0.0001

Data Index: 514604 (C4 (Web Text))

Max Activating Token Index: 851

Text #4

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2201. Min Act: -0.0001

Data Index: 1380033 (Python Code)

Max Activating Token Index: 182

Text #5

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2239. Min Act: -0.0001

Data Index: 674937 (C4 (Web Text))

Max Activating Token Index: 697

Text #6

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.2128. Min Act: -0.0001

Data Index: 1427408 (Python Code)

Max Activating Token Index: 17

Text #7

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1958. Min Act: -0.0001

Data Index: 933879 (C4 (Web Text))

Max Activating Token Index: 124

Text #8

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1989. Min Act: -0.0001

Data Index: 1046188 (C4 (Web Text))

Max Activating Token Index: 577

Text #9

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1967. Min Act: -0.0001

Data Index: 483948 (C4 (Web Text))

Max Activating Token Index: 64

Text #10

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1831. Min Act: -0.0001

Data Index: 1317213 (C4 (Web Text))

Max Activating Token Index: 204

Text #11

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1894. Min Act: -0.0001

Data Index: 1579195 (Python Code)

Max Activating Token Index: 383

Text #12

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1836. Min Act: -0.0001

Data Index: 248030 (C4 (Web Text))

Max Activating Token Index: 39

Text #13

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1805. Min Act: -0.0001

Data Index: 648422 (C4 (Web Text))

Max Activating Token Index: 439

Text #14

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1817. Min Act: -0.0001

Data Index: 1535534 (Python Code)

Max Activating Token Index: 669

Text #15

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1783. Min Act: -0.0001

Data Index: 767322 (C4 (Web Text))

Max Activating Token Index: 501

Text #16

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1826. Min Act: -0.0001

Data Index: 1452296 (Python Code)

Max Activating Token Index: 846

Text #17

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1741. Min Act: -0.0001

Data Index: 549598 (C4 (Web Text))

Max Activating Token Index: 106

Text #18

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1720. Min Act: -0.0001

Data Index: 1232218 (C4 (Web Text))

Max Activating Token Index: 585

Text #19

Max Range: 0.2604. Min Range: -0.2604

Max Act: 0.1711. Min Act: -0.0001

Data Index: 855669 (C4 (Web Text))

Max Activating Token Index: 660