Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 409 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0871. Min Act: -0.0003

Data Index: 1028954 (C4 (Web Text))

Max Activating Token Index: 288

Text #1

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0847. Min Act: -0.0003

Data Index: 1599938 (Python Code)

Max Activating Token Index: 456

Text #2

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0828. Min Act: -0.0003

Data Index: 345187 (C4 (Web Text))

Max Activating Token Index: 722

Text #3

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0790. Min Act: -0.0003

Data Index: 1342493 (C4 (Web Text))

Max Activating Token Index: 388

Text #4

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0812. Min Act: -0.0003

Data Index: 901313 (C4 (Web Text))

Max Activating Token Index: 713

Text #5

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0787. Min Act: -0.0003

Data Index: 1196327 (C4 (Web Text))

Max Activating Token Index: 80

Text #6

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0770. Min Act: -0.0003

Data Index: 1634387 (Python Code)

Max Activating Token Index: 707

Text #7

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0759. Min Act: -0.0002

Data Index: 577546 (C4 (Web Text))

Max Activating Token Index: 616

Text #8

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0745. Min Act: -0.0003

Data Index: 716181 (C4 (Web Text))

Max Activating Token Index: 677

Text #9

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0760. Min Act: -0.0003

Data Index: 1279237 (C4 (Web Text))

Max Activating Token Index: 643

Text #10

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0761. Min Act: -0.0003

Data Index: 1364142 (Python Code)

Max Activating Token Index: 981

Text #11

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0768. Min Act: -0.0003

Data Index: 131394 (C4 (Web Text))

Max Activating Token Index: 385

Text #12

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0744. Min Act: -0.0003

Data Index: 908937 (C4 (Web Text))

Max Activating Token Index: 65

Text #13

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0740. Min Act: -0.0004

Data Index: 494235 (C4 (Web Text))

Max Activating Token Index: 694

Text #14

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0770. Min Act: -0.0002

Data Index: 218071 (C4 (Web Text))

Max Activating Token Index: 69

Text #15

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0755. Min Act: -0.0003

Data Index: 1337346 (C4 (Web Text))

Max Activating Token Index: 117

Text #16

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0750. Min Act: -0.0003

Data Index: 521893 (C4 (Web Text))

Max Activating Token Index: 115

Text #17

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0741. Min Act: -0.0003

Data Index: 519376 (C4 (Web Text))

Max Activating Token Index: 131

Text #18

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0726. Min Act: -0.0002

Data Index: 1527238 (Python Code)

Max Activating Token Index: 601

Text #19

Max Range: 0.0871. Min Range: -0.0871

Max Act: 0.0740. Min Act: -0.0003

Data Index: 15118 (C4 (Web Text))

Max Activating Token Index: 523