Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 600 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6370. Min Act: -0.0000

Data Index: 883598 (C4 (Web Text))

Max Activating Token Index: 13

Text #1

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6138. Min Act: -0.0000

Data Index: 446719 (C4 (Web Text))

Max Activating Token Index: 691

Text #2

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6043. Min Act: -0.0000

Data Index: 758330 (C4 (Web Text))

Max Activating Token Index: 399

Text #3

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6022. Min Act: -0.0001

Data Index: 735241 (C4 (Web Text))

Max Activating Token Index: 7

Text #4

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6089. Min Act: -0.0000

Data Index: 48091 (C4 (Web Text))

Max Activating Token Index: 848

Text #5

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5944. Min Act: -0.0000

Data Index: 681670 (C4 (Web Text))

Max Activating Token Index: 969

Text #6

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5237. Min Act: -0.0000

Data Index: 727871 (C4 (Web Text))

Max Activating Token Index: 918

Text #7

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6121. Min Act: -0.0000

Data Index: 662996 (C4 (Web Text))

Max Activating Token Index: 51

Text #8

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6256. Min Act: -0.0001

Data Index: 1125955 (C4 (Web Text))

Max Activating Token Index: 507

Text #9

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6279. Min Act: -0.0000

Data Index: 398287 (C4 (Web Text))

Max Activating Token Index: 906

Text #10

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6015. Min Act: -0.0000

Data Index: 53296 (C4 (Web Text))

Max Activating Token Index: 7

Text #11

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5806. Min Act: -0.0000

Data Index: 822032 (C4 (Web Text))

Max Activating Token Index: 27

Text #12

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5920. Min Act: -0.0000

Data Index: 882810 (C4 (Web Text))

Max Activating Token Index: 144

Text #13

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.6046. Min Act: -0.0000

Data Index: 580527 (C4 (Web Text))

Max Activating Token Index: 251

Text #14

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5991. Min Act: -0.0000

Data Index: 795474 (C4 (Web Text))

Max Activating Token Index: 908

Text #15

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5769. Min Act: -0.0000

Data Index: 736948 (C4 (Web Text))

Max Activating Token Index: 693

Text #16

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5975. Min Act: -0.0000

Data Index: 944479 (C4 (Web Text))

Max Activating Token Index: 43

Text #17

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5928. Min Act: -0.0000

Data Index: 1325693 (C4 (Web Text))

Max Activating Token Index: 313

Text #18

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5605. Min Act: -0.0000

Data Index: 494914 (C4 (Web Text))

Max Activating Token Index: 475

Text #19

Max Range: 0.6370. Min Range: -0.6370

Max Act: 0.5569. Min Act: -0.0001

Data Index: 1097443 (C4 (Web Text))

Max Activating Token Index: 657