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

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

Neuron 5154 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.3085. Min Act: -0.0000

Data Index: 1118815 (C4 (Web Text))

Max Activating Token Index: 318

Text #1

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.3841. Min Act: -0.0000

Data Index: 380244 (C4 (Web Text))

Max Activating Token Index: 677

Text #2

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.3572. Min Act: -0.0000

Data Index: 223663 (C4 (Web Text))

Max Activating Token Index: 300

Text #3

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.5150. Min Act: -0.0000

Data Index: 58662 (C4 (Web Text))

Max Activating Token Index: 984

Text #4

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.2306. Min Act: -0.0000

Data Index: 391024 (C4 (Web Text))

Max Activating Token Index: 744

Text #5

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.2439. Min Act: -0.0000

Data Index: 881395 (C4 (Web Text))

Max Activating Token Index: 966

Text #6

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.2137. Min Act: -0.0000

Data Index: 840745 (C4 (Web Text))

Max Activating Token Index: 569

Text #7

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.2823. Min Act: -0.0000

Data Index: 1132063 (C4 (Web Text))

Max Activating Token Index: 357

Text #8

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.1517. Min Act: -0.0000

Data Index: 69692 (C4 (Web Text))

Max Activating Token Index: 392

Text #9

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0378. Min Act: -0.0000

Data Index: 871007 (C4 (Web Text))

Max Activating Token Index: 569

Text #10

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.1113. Min Act: -0.0000

Data Index: 78570 (C4 (Web Text))

Max Activating Token Index: 40

Text #11

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.1073. Min Act: -0.0000

Data Index: 993528 (C4 (Web Text))

Max Activating Token Index: 1006

Text #12

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0639. Min Act: -0.0000

Data Index: 1383448 (Python Code)

Max Activating Token Index: 833

Text #13

Max Range: 2.5150. Min Range: -2.5150

Max Act: 1.9662. Min Act: -0.0000

Data Index: 5868 (C4 (Web Text))

Max Activating Token Index: 769

Text #14

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0370. Min Act: -0.0000

Data Index: 583791 (C4 (Web Text))

Max Activating Token Index: 1002

Text #15

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0334. Min Act: -0.0000

Data Index: 206626 (C4 (Web Text))

Max Activating Token Index: 919

Text #16

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0359. Min Act: -0.0000

Data Index: 381914 (C4 (Web Text))

Max Activating Token Index: 467

Text #17

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0186. Min Act: -0.0000

Data Index: 184790 (C4 (Web Text))

Max Activating Token Index: 739

Text #18

Max Range: 2.5150. Min Range: -2.5150

Max Act: 1.9978. Min Act: -0.0000

Data Index: 99580 (C4 (Web Text))

Max Activating Token Index: 668

Text #19

Max Range: 2.5150. Min Range: -2.5150

Max Act: 2.0844. Min Act: -0.0000

Data Index: 433742 (C4 (Web Text))

Max Activating Token Index: 235