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

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

Neuron 3228 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.2749. Min Act: -0.0000

Data Index: 165533 (C4 (Web Text))

Max Activating Token Index: 544

Text #1

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.2216. Min Act: -0.0000

Data Index: 1418600 (Python Code)

Max Activating Token Index: 475

Text #2

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.2446. Min Act: -0.0000

Data Index: 1467465 (Python Code)

Max Activating Token Index: 355

Text #3

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.2116. Min Act: -0.0000

Data Index: 213623 (C4 (Web Text))

Max Activating Token Index: 994

Text #4

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.2089. Min Act: -0.0000

Data Index: 298668 (C4 (Web Text))

Max Activating Token Index: 852

Text #5

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1884. Min Act: -0.0000

Data Index: 990516 (C4 (Web Text))

Max Activating Token Index: 585

Text #6

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1921. Min Act: -0.0000

Data Index: 1483954 (Python Code)

Max Activating Token Index: 856

Text #7

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1945. Min Act: 0.0000

Data Index: 161212 (C4 (Web Text))

Max Activating Token Index: 173

Text #8

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1792. Min Act: -0.0000

Data Index: 1188181 (C4 (Web Text))

Max Activating Token Index: 260

Text #9

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1847. Min Act: -0.0000

Data Index: 1029705 (C4 (Web Text))

Max Activating Token Index: 961

Text #10

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1855. Min Act: -0.0000

Data Index: 1532406 (Python Code)

Max Activating Token Index: 17

Text #11

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1898. Min Act: -0.0000

Data Index: 1556791 (Python Code)

Max Activating Token Index: 68

Text #12

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1817. Min Act: -0.0000

Data Index: 307525 (C4 (Web Text))

Max Activating Token Index: 1012

Text #13

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1789. Min Act: -0.0000

Data Index: 902389 (C4 (Web Text))

Max Activating Token Index: 884

Text #14

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1746. Min Act: -0.0000

Data Index: 1656122 (Python Code)

Max Activating Token Index: 340

Text #15

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1844. Min Act: -0.0000

Data Index: 1325376 (C4 (Web Text))

Max Activating Token Index: 988

Text #16

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1720. Min Act: -0.0000

Data Index: 524596 (C4 (Web Text))

Max Activating Token Index: 419

Text #17

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1768. Min Act: -0.0000

Data Index: 1045470 (C4 (Web Text))

Max Activating Token Index: 556

Text #18

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1565. Min Act: -0.0000

Data Index: 792463 (C4 (Web Text))

Max Activating Token Index: 798

Text #19

Max Range: 0.2749. Min Range: -0.2749

Max Act: 0.1696. Min Act: -0.0000

Data Index: 1645051 (Python Code)

Max Activating Token Index: 735