Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 58 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0719. Min Act: -0.0001

Data Index: 386396 (C4 (Web Text))

Max Activating Token Index: 732

Text #1

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0672. Min Act: -0.0001

Data Index: 92215 (C4 (Web Text))

Max Activating Token Index: 290

Text #2

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0674. Min Act: -0.0001

Data Index: 909205 (C4 (Web Text))

Max Activating Token Index: 266

Text #3

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0679. Min Act: -0.0002

Data Index: 1135148 (C4 (Web Text))

Max Activating Token Index: 274

Text #4

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0605. Min Act: -0.0001

Data Index: 31906 (C4 (Web Text))

Max Activating Token Index: 794

Text #5

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0675. Min Act: -0.0002

Data Index: 90979 (C4 (Web Text))

Max Activating Token Index: 264

Text #6

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0648. Min Act: -0.0002

Data Index: 1166112 (C4 (Web Text))

Max Activating Token Index: 918

Text #7

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0633. Min Act: -0.0002

Data Index: 324508 (C4 (Web Text))

Max Activating Token Index: 659

Text #8

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0577. Min Act: -0.0002

Data Index: 280612 (C4 (Web Text))

Max Activating Token Index: 746

Text #9

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0636. Min Act: -0.0001

Data Index: 1300970 (C4 (Web Text))

Max Activating Token Index: 534

Text #10

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0643. Min Act: -0.0001

Data Index: 985974 (C4 (Web Text))

Max Activating Token Index: 364

Text #11

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0599. Min Act: -0.0001

Data Index: 972065 (C4 (Web Text))

Max Activating Token Index: 194

Text #12

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0592. Min Act: -0.0002

Data Index: 814677 (C4 (Web Text))

Max Activating Token Index: 691

Text #13

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0574. Min Act: -0.0001

Data Index: 1227908 (C4 (Web Text))

Max Activating Token Index: 730

Text #14

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0573. Min Act: -0.0001

Data Index: 1640440 (Python Code)

Max Activating Token Index: 847

Text #15

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0556. Min Act: -0.0001

Data Index: 1496484 (Python Code)

Max Activating Token Index: 104

Text #16

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0605. Min Act: -0.0001

Data Index: 1124920 (C4 (Web Text))

Max Activating Token Index: 307

Text #17

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0560. Min Act: -0.0001

Data Index: 621816 (C4 (Web Text))

Max Activating Token Index: 938

Text #18

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0541. Min Act: -0.0002

Data Index: 424768 (C4 (Web Text))

Max Activating Token Index: 464

Text #19

Max Range: 0.0719. Min Range: -0.0719

Max Act: 0.0535. Min Act: -0.0001

Data Index: 574092 (C4 (Web Text))

Max Activating Token Index: 982