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

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

Neuron 3053 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0704. Min Act: -0.0002

Data Index: 1653958 (Python Code)

Max Activating Token Index: 433

Text #1

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0590. Min Act: -0.0002

Data Index: 68419 (C4 (Web Text))

Max Activating Token Index: 724

Text #2

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0679. Min Act: -0.0002

Data Index: 1013854 (C4 (Web Text))

Max Activating Token Index: 340

Text #3

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0686. Min Act: -0.0002

Data Index: 243135 (C4 (Web Text))

Max Activating Token Index: 764

Text #4

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0652. Min Act: -0.0002

Data Index: 701377 (C4 (Web Text))

Max Activating Token Index: 130

Text #5

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0614. Min Act: -0.0002

Data Index: 729183 (C4 (Web Text))

Max Activating Token Index: 496

Text #6

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0709. Min Act: -0.0002

Data Index: 361452 (C4 (Web Text))

Max Activating Token Index: 316

Text #7

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0633. Min Act: -0.0002

Data Index: 129904 (C4 (Web Text))

Max Activating Token Index: 49

Text #8

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0729. Min Act: -0.0002

Data Index: 385396 (C4 (Web Text))

Max Activating Token Index: 152

Text #9

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0613. Min Act: -0.0002

Data Index: 1635292 (Python Code)

Max Activating Token Index: 65

Text #10

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0644. Min Act: -0.0002

Data Index: 1277671 (C4 (Web Text))

Max Activating Token Index: 479

Text #11

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0624. Min Act: -0.0002

Data Index: 468162 (C4 (Web Text))

Max Activating Token Index: 366

Text #12

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0642. Min Act: -0.0002

Data Index: 1321893 (C4 (Web Text))

Max Activating Token Index: 569

Text #13

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0683. Min Act: -0.0002

Data Index: 1135232 (C4 (Web Text))

Max Activating Token Index: 17

Text #14

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0554. Min Act: -0.0002

Data Index: 801175 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0580. Min Act: -0.0002

Data Index: 727928 (C4 (Web Text))

Max Activating Token Index: 938

Text #16

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0601. Min Act: -0.0002

Data Index: 1456355 (Python Code)

Max Activating Token Index: 74

Text #17

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0609. Min Act: -0.0002

Data Index: 1036629 (C4 (Web Text))

Max Activating Token Index: 278

Text #18

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0612. Min Act: -0.0002

Data Index: 395390 (C4 (Web Text))

Max Activating Token Index: 896

Text #19

Max Range: 0.0729. Min Range: -0.0729

Max Act: 0.0615. Min Act: -0.0002

Data Index: 1344226 (C4 (Web Text))

Max Activating Token Index: 54