Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1596 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1719. Min Act: -0.0001

Data Index: 653020 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1767. Min Act: -0.0001

Data Index: 1148751 (C4 (Web Text))

Max Activating Token Index: 773

Text #2

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1660. Min Act: -0.0001

Data Index: 1044862 (C4 (Web Text))

Max Activating Token Index: 678

Text #3

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1717. Min Act: -0.0001

Data Index: 471589 (C4 (Web Text))

Max Activating Token Index: 821

Text #4

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1694. Min Act: -0.0001

Data Index: 1288413 (C4 (Web Text))

Max Activating Token Index: 155

Text #5

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1702. Min Act: -0.0001

Data Index: 160493 (C4 (Web Text))

Max Activating Token Index: 186

Text #6

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1707. Min Act: -0.0001

Data Index: 1022445 (C4 (Web Text))

Max Activating Token Index: 685

Text #7

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1571. Min Act: -0.0001

Data Index: 618264 (C4 (Web Text))

Max Activating Token Index: 627

Text #8

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1709. Min Act: -0.0001

Data Index: 15055 (C4 (Web Text))

Max Activating Token Index: 43

Text #9

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1700. Min Act: -0.0001

Data Index: 643987 (C4 (Web Text))

Max Activating Token Index: 68

Text #10

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1635. Min Act: -0.0001

Data Index: 1287056 (C4 (Web Text))

Max Activating Token Index: 178

Text #11

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1586. Min Act: -0.0001

Data Index: 233290 (C4 (Web Text))

Max Activating Token Index: 786

Text #12

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1546. Min Act: -0.0001

Data Index: 1172096 (C4 (Web Text))

Max Activating Token Index: 602

Text #13

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1551. Min Act: -0.0001

Data Index: 564891 (C4 (Web Text))

Max Activating Token Index: 103

Text #14

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1677. Min Act: -0.0001

Data Index: 364921 (C4 (Web Text))

Max Activating Token Index: 387

Text #15

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1596. Min Act: -0.0001

Data Index: 197649 (C4 (Web Text))

Max Activating Token Index: 532

Text #16

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1470. Min Act: -0.0001

Data Index: 150741 (C4 (Web Text))

Max Activating Token Index: 876

Text #17

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1585. Min Act: -0.0001

Data Index: 465129 (C4 (Web Text))

Max Activating Token Index: 150

Text #18

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1638. Min Act: -0.0001

Data Index: 380028 (C4 (Web Text))

Max Activating Token Index: 196

Text #19

Max Range: 0.1767. Min Range: -0.1767

Max Act: 0.1580. Min Act: -0.0001

Data Index: 16367 (C4 (Web Text))

Max Activating Token Index: 132