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

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

Neuron 4617 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4730. Min Range: -5.4730

Max Act: 5.4730. Min Act: -0.0000

Data Index: 1625414 (Python Code)

Max Activating Token Index: 212

Text #1

Max Range: 5.4730. Min Range: -5.4730

Max Act: 5.3680. Min Act: -0.0000

Data Index: 1535598 (Python Code)

Max Activating Token Index: 108

Text #2

Max Range: 5.4730. Min Range: -5.4730

Max Act: 5.1154. Min Act: -0.0000

Data Index: 1494210 (Python Code)

Max Activating Token Index: 751

Text #3

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.7356. Min Act: -0.0000

Data Index: 1462240 (Python Code)

Max Activating Token Index: 889

Text #4

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.4609. Min Act: -0.0000

Data Index: 1633751 (Python Code)

Max Activating Token Index: 347

Text #5

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.0855. Min Act: -0.0000

Data Index: 1179420 (C4 (Web Text))

Max Activating Token Index: 583

Text #6

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.1729. Min Act: -0.0000

Data Index: 1384764 (Python Code)

Max Activating Token Index: 849

Text #7

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.7190. Min Act: -0.0000

Data Index: 1244394 (C4 (Web Text))

Max Activating Token Index: 1013

Text #8

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.0344. Min Act: -0.0000

Data Index: 1548317 (Python Code)

Max Activating Token Index: 234

Text #9

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.1518. Min Act: -0.0000

Data Index: 493392 (C4 (Web Text))

Max Activating Token Index: 234

Text #10

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.8129. Min Act: -0.0000

Data Index: 1513664 (Python Code)

Max Activating Token Index: 66

Text #11

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.7477. Min Act: -0.0000

Data Index: 46138 (C4 (Web Text))

Max Activating Token Index: 732

Text #12

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.9637. Min Act: -0.0000

Data Index: 1253107 (C4 (Web Text))

Max Activating Token Index: 673

Text #13

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.7862. Min Act: -0.0000

Data Index: 332733 (C4 (Web Text))

Max Activating Token Index: 250

Text #14

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.0825. Min Act: -0.0000

Data Index: 1314423 (C4 (Web Text))

Max Activating Token Index: 710

Text #15

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.6982. Min Act: -0.0000

Data Index: 391596 (C4 (Web Text))

Max Activating Token Index: 961

Text #16

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.8046. Min Act: -0.0000

Data Index: 1548790 (Python Code)

Max Activating Token Index: 996

Text #17

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.4946. Min Act: -0.0000

Data Index: 1543764 (Python Code)

Max Activating Token Index: 151

Text #18

Max Range: 5.4730. Min Range: -5.4730

Max Act: 3.8389. Min Act: -0.0000

Data Index: 655828 (C4 (Web Text))

Max Activating Token Index: 636

Text #19

Max Range: 5.4730. Min Range: -5.4730

Max Act: 4.0470. Min Act: -0.0000

Data Index: 402167 (C4 (Web Text))

Max Activating Token Index: 999