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

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

Neuron 4512 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7789. Min Range: -2.7789

Max Act: 2.7789. Min Act: -0.0000

Data Index: 1198879 (C4 (Web Text))

Max Activating Token Index: 413

Text #1

Max Range: 2.7789. Min Range: -2.7789

Max Act: 2.6362. Min Act: -0.0000

Data Index: 811114 (C4 (Web Text))

Max Activating Token Index: 295

Text #2

Max Range: 2.7789. Min Range: -2.7789

Max Act: 2.4516. Min Act: -0.0001

Data Index: 1599721 (Python Code)

Max Activating Token Index: 627

Text #3

Max Range: 2.7789. Min Range: -2.7789

Max Act: 2.0398. Min Act: -0.0001

Data Index: 1533951 (Python Code)

Max Activating Token Index: 22

Text #4

Max Range: 2.7789. Min Range: -2.7789

Max Act: 2.1199. Min Act: -0.0000

Data Index: 1348942 (C4 (Web Text))

Max Activating Token Index: 126

Text #5

Max Range: 2.7789. Min Range: -2.7789

Max Act: 2.1069. Min Act: -0.0000

Data Index: 38661 (C4 (Web Text))

Max Activating Token Index: 879

Text #6

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.9809. Min Act: -0.0000

Data Index: 1474089 (Python Code)

Max Activating Token Index: 827

Text #7

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.9745. Min Act: -0.0000

Data Index: 1469856 (Python Code)

Max Activating Token Index: 52

Text #8

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.8774. Min Act: -0.0000

Data Index: 1516851 (Python Code)

Max Activating Token Index: 561

Text #9

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.9506. Min Act: -0.0000

Data Index: 1291200 (C4 (Web Text))

Max Activating Token Index: 631

Text #10

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.8409. Min Act: -0.0000

Data Index: 1579219 (Python Code)

Max Activating Token Index: 397

Text #11

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.8380. Min Act: -0.0000

Data Index: 1533724 (Python Code)

Max Activating Token Index: 791

Text #12

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.7547. Min Act: -0.0000

Data Index: 1387945 (Python Code)

Max Activating Token Index: 758

Text #13

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.7009. Min Act: -0.0001

Data Index: 1399171 (Python Code)

Max Activating Token Index: 195

Text #14

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.7359. Min Act: -0.0000

Data Index: 827446 (C4 (Web Text))

Max Activating Token Index: 927

Text #15

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.8799. Min Act: -0.0000

Data Index: 1470051 (Python Code)

Max Activating Token Index: 213

Text #16

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.8805. Min Act: -0.0001

Data Index: 1433176 (Python Code)

Max Activating Token Index: 343

Text #17

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.7504. Min Act: -0.0000

Data Index: 1558167 (Python Code)

Max Activating Token Index: 293

Text #18

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.7063. Min Act: -0.0000

Data Index: 1620298 (Python Code)

Max Activating Token Index: 604

Text #19

Max Range: 2.7789. Min Range: -2.7789

Max Act: 1.8472. Min Act: -0.0000

Data Index: 1417539 (Python Code)

Max Activating Token Index: 751