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

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

Neuron 3387 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5050. Min Act: -0.0000

Data Index: 1594201 (Python Code)

Max Activating Token Index: 924

Text #1

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.6778. Min Act: -0.0000

Data Index: 1506300 (Python Code)

Max Activating Token Index: 553

Text #2

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5978. Min Act: -0.0000

Data Index: 57126 (C4 (Web Text))

Max Activating Token Index: 722

Text #3

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.6045. Min Act: -0.0000

Data Index: 1578261 (Python Code)

Max Activating Token Index: 766

Text #4

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.6318. Min Act: -0.0000

Data Index: 1471306 (Python Code)

Max Activating Token Index: 620

Text #5

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5798. Min Act: -0.0000

Data Index: 1640012 (Python Code)

Max Activating Token Index: 945

Text #6

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.6472. Min Act: -0.0000

Data Index: 1444359 (Python Code)

Max Activating Token Index: 648

Text #7

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5518. Min Act: -0.0000

Data Index: 43298 (C4 (Web Text))

Max Activating Token Index: 531

Text #8

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5851. Min Act: -0.0000

Data Index: 261869 (C4 (Web Text))

Max Activating Token Index: 472

Text #9

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5824. Min Act: -0.0000

Data Index: 1583738 (Python Code)

Max Activating Token Index: 317

Text #10

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5173. Min Act: -0.0000

Data Index: 1538235 (Python Code)

Max Activating Token Index: 985

Text #11

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5064. Min Act: -0.0000

Data Index: 1365543 (Python Code)

Max Activating Token Index: 564

Text #12

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5554. Min Act: -0.0000

Data Index: 1373111 (Python Code)

Max Activating Token Index: 323

Text #13

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5373. Min Act: -0.0000

Data Index: 1584471 (Python Code)

Max Activating Token Index: 452

Text #14

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.4122. Min Act: -0.0000

Data Index: 1398632 (Python Code)

Max Activating Token Index: 100

Text #15

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.4735. Min Act: -0.0000

Data Index: 1450410 (Python Code)

Max Activating Token Index: 671

Text #16

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5130. Min Act: -0.0000

Data Index: 1391139 (Python Code)

Max Activating Token Index: 264

Text #17

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.4980. Min Act: -0.0000

Data Index: 2610 (C4 (Web Text))

Max Activating Token Index: 63

Text #18

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.4276. Min Act: -0.0000

Data Index: 1377694 (Python Code)

Max Activating Token Index: 143

Text #19

Max Range: 0.6778. Min Range: -0.6778

Max Act: 0.5927. Min Act: -0.0000

Data Index: 164496 (C4 (Web Text))

Max Activating Token Index: 174