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

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

Neuron 298 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: 0.2522. Min Range: -0.2522

Max Act: 0.2336. Min Act: -0.0000

Data Index: 320188 (C4 (Web Text))

Max Activating Token Index: 259

Text #1

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.2522. Min Act: -0.0000

Data Index: 1012163 (C4 (Web Text))

Max Activating Token Index: 337

Text #2

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.2333. Min Act: -0.0000

Data Index: 1173343 (C4 (Web Text))

Max Activating Token Index: 36

Text #3

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.2056. Min Act: -0.0000

Data Index: 260334 (C4 (Web Text))

Max Activating Token Index: 227

Text #4

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1988. Min Act: -0.0000

Data Index: 602340 (C4 (Web Text))

Max Activating Token Index: 993

Text #5

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1919. Min Act: -0.0000

Data Index: 1529588 (Python Code)

Max Activating Token Index: 51

Text #6

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1884. Min Act: -0.0000

Data Index: 1357063 (C4 (Web Text))

Max Activating Token Index: 800

Text #7

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.2013. Min Act: -0.0000

Data Index: 1655310 (Python Code)

Max Activating Token Index: 69

Text #8

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1909. Min Act: -0.0000

Data Index: 1482841 (Python Code)

Max Activating Token Index: 918

Text #9

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1982. Min Act: -0.0000

Data Index: 562131 (C4 (Web Text))

Max Activating Token Index: 119

Text #10

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1853. Min Act: -0.0000

Data Index: 760682 (C4 (Web Text))

Max Activating Token Index: 584

Text #11

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1886. Min Act: -0.0000

Data Index: 1590528 (Python Code)

Max Activating Token Index: 55

Text #12

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1859. Min Act: -0.0000

Data Index: 362032 (C4 (Web Text))

Max Activating Token Index: 950

Text #13

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1824. Min Act: -0.0000

Data Index: 566536 (C4 (Web Text))

Max Activating Token Index: 926

Text #14

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1674. Min Act: -0.0000

Data Index: 1446617 (Python Code)

Max Activating Token Index: 649

Text #15

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1697. Min Act: -0.0000

Data Index: 1378910 (Python Code)

Max Activating Token Index: 214

Text #16

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1589. Min Act: -0.0000

Data Index: 678105 (C4 (Web Text))

Max Activating Token Index: 924

Text #17

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1569. Min Act: -0.0000

Data Index: 76493 (C4 (Web Text))

Max Activating Token Index: 161

Text #18

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1689. Min Act: -0.0000

Data Index: 1502794 (Python Code)

Max Activating Token Index: 613

Text #19

Max Range: 0.2522. Min Range: -0.2522

Max Act: 0.1742. Min Act: -0.0000

Data Index: 919734 (C4 (Web Text))

Max Activating Token Index: 178