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

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

Neuron 4804 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5285. Min Range: -3.5285

Max Act: 3.5285. Min Act: -0.0001

Data Index: 916714 (C4 (Web Text))

Max Activating Token Index: 609

Text #1

Max Range: 3.5285. Min Range: -3.5285

Max Act: 3.2566. Min Act: -0.0001

Data Index: 326993 (C4 (Web Text))

Max Activating Token Index: 482

Text #2

Max Range: 3.5285. Min Range: -3.5285

Max Act: 3.1879. Min Act: -0.0001

Data Index: 132704 (C4 (Web Text))

Max Activating Token Index: 70

Text #3

Max Range: 3.5285. Min Range: -3.5285

Max Act: 3.2504. Min Act: -0.0001

Data Index: 46698 (C4 (Web Text))

Max Activating Token Index: 559

Text #4

Max Range: 3.5285. Min Range: -3.5285

Max Act: 3.4005. Min Act: -0.0001

Data Index: 433496 (C4 (Web Text))

Max Activating Token Index: 218

Text #5

Max Range: 3.5285. Min Range: -3.5285

Max Act: 3.0041. Min Act: -0.0001

Data Index: 142600 (C4 (Web Text))

Max Activating Token Index: 386

Text #6

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.9940. Min Act: -0.0000

Data Index: 412063 (C4 (Web Text))

Max Activating Token Index: 485

Text #7

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.9325. Min Act: -0.0001

Data Index: 692783 (C4 (Web Text))

Max Activating Token Index: 212

Text #8

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.8975. Min Act: -0.0000

Data Index: 237784 (C4 (Web Text))

Max Activating Token Index: 23

Text #9

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.8706. Min Act: -0.0000

Data Index: 865050 (C4 (Web Text))

Max Activating Token Index: 70

Text #10

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.8541. Min Act: -0.0001

Data Index: 518190 (C4 (Web Text))

Max Activating Token Index: 696

Text #11

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.8013. Min Act: -0.0001

Data Index: 922268 (C4 (Web Text))

Max Activating Token Index: 11

Text #12

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.8009. Min Act: -0.0001

Data Index: 115296 (C4 (Web Text))

Max Activating Token Index: 588

Text #13

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.5784. Min Act: -0.0001

Data Index: 1035403 (C4 (Web Text))

Max Activating Token Index: 21

Text #14

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.6895. Min Act: -0.0001

Data Index: 940733 (C4 (Web Text))

Max Activating Token Index: 425

Text #15

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.7958. Min Act: -0.0001

Data Index: 1049604 (C4 (Web Text))

Max Activating Token Index: 366

Text #16

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.6664. Min Act: -0.0001

Data Index: 1184923 (C4 (Web Text))

Max Activating Token Index: 24

Text #17

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.4827. Min Act: -0.0001

Data Index: 938377 (C4 (Web Text))

Max Activating Token Index: 443

Text #18

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.6122. Min Act: -0.0001

Data Index: 1059843 (C4 (Web Text))

Max Activating Token Index: 39

Text #19

Max Range: 3.5285. Min Range: -3.5285

Max Act: 2.4820. Min Act: -0.0000

Data Index: 61625 (C4 (Web Text))

Max Activating Token Index: 750