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

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

Neuron 4105 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1847. Min Act: -0.0001

Data Index: 188439 (C4 (Web Text))

Max Activating Token Index: 497

Text #1

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1974. Min Act: -0.0000

Data Index: 835091 (C4 (Web Text))

Max Activating Token Index: 447

Text #2

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1788. Min Act: -0.0001

Data Index: 270818 (C4 (Web Text))

Max Activating Token Index: 993

Text #3

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1742. Min Act: -0.0001

Data Index: 824820 (C4 (Web Text))

Max Activating Token Index: 324

Text #4

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1703. Min Act: -0.0001

Data Index: 1466654 (Python Code)

Max Activating Token Index: 43

Text #5

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1614. Min Act: -0.0001

Data Index: 67479 (C4 (Web Text))

Max Activating Token Index: 962

Text #6

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1650. Min Act: -0.0001

Data Index: 418280 (C4 (Web Text))

Max Activating Token Index: 570

Text #7

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1693. Min Act: -0.0000

Data Index: 1370407 (Python Code)

Max Activating Token Index: 535

Text #8

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1631. Min Act: -0.0000

Data Index: 1421795 (Python Code)

Max Activating Token Index: 60

Text #9

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1665. Min Act: -0.0001

Data Index: 738290 (C4 (Web Text))

Max Activating Token Index: 778

Text #10

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1545. Min Act: -0.0001

Data Index: 1541326 (Python Code)

Max Activating Token Index: 483

Text #11

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1707. Min Act: -0.0001

Data Index: 1415648 (Python Code)

Max Activating Token Index: 39

Text #12

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1584. Min Act: -0.0001

Data Index: 1418883 (Python Code)

Max Activating Token Index: 699

Text #13

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1579. Min Act: -0.0000

Data Index: 475466 (C4 (Web Text))

Max Activating Token Index: 330

Text #14

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1472. Min Act: -0.0001

Data Index: 1612624 (Python Code)

Max Activating Token Index: 36

Text #15

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1507. Min Act: -0.0001

Data Index: 1555357 (Python Code)

Max Activating Token Index: 890

Text #16

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1501. Min Act: -0.0001

Data Index: 1027133 (C4 (Web Text))

Max Activating Token Index: 942

Text #17

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1536. Min Act: -0.0001

Data Index: 1566568 (Python Code)

Max Activating Token Index: 401

Text #18

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1594. Min Act: -0.0000

Data Index: 1540874 (Python Code)

Max Activating Token Index: 961

Text #19

Max Range: 0.1974. Min Range: -0.1974

Max Act: 0.1521. Min Act: -0.0000

Data Index: 913640 (C4 (Web Text))

Max Activating Token Index: 568