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

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

Neuron 1563 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.1069. Min Act: -0.0001

Data Index: 944122 (C4 (Web Text))

Max Activating Token Index: 4

Text #1

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0953. Min Act: -0.0001

Data Index: 11021 (C4 (Web Text))

Max Activating Token Index: 815

Text #2

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0760. Min Act: -0.0001

Data Index: 1598025 (Python Code)

Max Activating Token Index: 169

Text #3

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0629. Min Act: -0.0001

Data Index: 888322 (C4 (Web Text))

Max Activating Token Index: 873

Text #4

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0620. Min Act: -0.0001

Data Index: 577624 (C4 (Web Text))

Max Activating Token Index: 906

Text #5

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0679. Min Act: -0.0001

Data Index: 1323904 (C4 (Web Text))

Max Activating Token Index: 996

Text #6

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0660. Min Act: -0.0001

Data Index: 1244486 (C4 (Web Text))

Max Activating Token Index: 802

Text #7

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0642. Min Act: -0.0001

Data Index: 361207 (C4 (Web Text))

Max Activating Token Index: 3

Text #8

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0525. Min Act: -0.0001

Data Index: 1594586 (Python Code)

Max Activating Token Index: 543

Text #9

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0434. Min Act: -0.0001

Data Index: 1544044 (Python Code)

Max Activating Token Index: 361

Text #10

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0553. Min Act: -0.0001

Data Index: 1248254 (C4 (Web Text))

Max Activating Token Index: 230

Text #11

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0535. Min Act: -0.0001

Data Index: 394564 (C4 (Web Text))

Max Activating Token Index: 156

Text #12

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0548. Min Act: -0.0001

Data Index: 1447102 (Python Code)

Max Activating Token Index: 706

Text #13

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0594. Min Act: -0.0001

Data Index: 87893 (C4 (Web Text))

Max Activating Token Index: 14

Text #14

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0473. Min Act: -0.0001

Data Index: 1451929 (Python Code)

Max Activating Token Index: 446

Text #15

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0580. Min Act: -0.0001

Data Index: 26570 (C4 (Web Text))

Max Activating Token Index: 823

Text #16

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0539. Min Act: -0.0001

Data Index: 215608 (C4 (Web Text))

Max Activating Token Index: 952

Text #17

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0539. Min Act: -0.0001

Data Index: 993685 (C4 (Web Text))

Max Activating Token Index: 5

Text #18

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0533. Min Act: -0.0001

Data Index: 1575930 (Python Code)

Max Activating Token Index: 164

Text #19

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0449. Min Act: -0.0001

Data Index: 211171 (C4 (Web Text))

Max Activating Token Index: 126