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

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

Neuron 2640 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.7730. Min Act: -0.0003

Data Index: 1331313 (C4 (Web Text))

Max Activating Token Index: 713

Text #1

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.7276. Min Act: -0.0003

Data Index: 540603 (C4 (Web Text))

Max Activating Token Index: 681

Text #2

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6911. Min Act: -0.0003

Data Index: 98280 (C4 (Web Text))

Max Activating Token Index: 945

Text #3

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6629. Min Act: -0.0002

Data Index: 1245931 (C4 (Web Text))

Max Activating Token Index: 697

Text #4

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6443. Min Act: -0.0002

Data Index: 1263814 (C4 (Web Text))

Max Activating Token Index: 158

Text #5

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6135. Min Act: -0.0003

Data Index: 824693 (C4 (Web Text))

Max Activating Token Index: 37

Text #6

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6198. Min Act: -0.0003

Data Index: 924453 (C4 (Web Text))

Max Activating Token Index: 911

Text #7

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5971. Min Act: -0.0003

Data Index: 84258 (C4 (Web Text))

Max Activating Token Index: 616

Text #8

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6069. Min Act: -0.0003

Data Index: 762110 (C4 (Web Text))

Max Activating Token Index: 206

Text #9

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6250. Min Act: -0.0003

Data Index: 1278845 (C4 (Web Text))

Max Activating Token Index: 1017

Text #10

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.6205. Min Act: -0.0003

Data Index: 751223 (C4 (Web Text))

Max Activating Token Index: 569

Text #11

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5776. Min Act: -0.0003

Data Index: 1065097 (C4 (Web Text))

Max Activating Token Index: 623

Text #12

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5980. Min Act: -0.0002

Data Index: 372113 (C4 (Web Text))

Max Activating Token Index: 757

Text #13

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5889. Min Act: -0.0002

Data Index: 465817 (C4 (Web Text))

Max Activating Token Index: 178

Text #14

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5870. Min Act: -0.0003

Data Index: 1300840 (C4 (Web Text))

Max Activating Token Index: 802

Text #15

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5770. Min Act: -0.0003

Data Index: 614294 (C4 (Web Text))

Max Activating Token Index: 26

Text #16

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5540. Min Act: -0.0003

Data Index: 251345 (C4 (Web Text))

Max Activating Token Index: 828

Text #17

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5679. Min Act: -0.0003

Data Index: 1352403 (C4 (Web Text))

Max Activating Token Index: 748

Text #18

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5619. Min Act: -0.0003

Data Index: 840409 (C4 (Web Text))

Max Activating Token Index: 141

Text #19

Max Range: 0.7730. Min Range: -0.7730

Max Act: 0.5528. Min Act: -0.0002

Data Index: 459826 (C4 (Web Text))

Max Activating Token Index: 635