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

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

Neuron 478 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.8420. Min Act: -0.0002

Data Index: 681804 (C4 (Web Text))

Max Activating Token Index: 925

Text #1

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.8195. Min Act: -0.0002

Data Index: 897715 (C4 (Web Text))

Max Activating Token Index: 44

Text #2

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7654. Min Act: -0.0002

Data Index: 1317798 (C4 (Web Text))

Max Activating Token Index: 31

Text #3

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6576. Min Act: -0.0002

Data Index: 666349 (C4 (Web Text))

Max Activating Token Index: 102

Text #4

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7108. Min Act: -0.0002

Data Index: 601218 (C4 (Web Text))

Max Activating Token Index: 35

Text #5

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7030. Min Act: -0.0002

Data Index: 1075539 (C4 (Web Text))

Max Activating Token Index: 467

Text #6

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7164. Min Act: -0.0002

Data Index: 560828 (C4 (Web Text))

Max Activating Token Index: 63

Text #7

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6267. Min Act: -0.0002

Data Index: 237332 (C4 (Web Text))

Max Activating Token Index: 887

Text #8

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7100. Min Act: -0.0002

Data Index: 184757 (C4 (Web Text))

Max Activating Token Index: 392

Text #9

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6865. Min Act: -0.0002

Data Index: 236693 (C4 (Web Text))

Max Activating Token Index: 123

Text #10

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7248. Min Act: -0.0002

Data Index: 95333 (C4 (Web Text))

Max Activating Token Index: 897

Text #11

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7001. Min Act: -0.0002

Data Index: 1561088 (Python Code)

Max Activating Token Index: 6

Text #12

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7148. Min Act: -0.0002

Data Index: 1156729 (C4 (Web Text))

Max Activating Token Index: 565

Text #13

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6266. Min Act: -0.0002

Data Index: 603405 (C4 (Web Text))

Max Activating Token Index: 949

Text #14

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6912. Min Act: -0.0002

Data Index: 506235 (C4 (Web Text))

Max Activating Token Index: 1016

Text #15

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6481. Min Act: -0.0002

Data Index: 155048 (C4 (Web Text))

Max Activating Token Index: 534

Text #16

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6883. Min Act: -0.0002

Data Index: 1139710 (C4 (Web Text))

Max Activating Token Index: 15

Text #17

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.7199. Min Act: -0.0002

Data Index: 371587 (C4 (Web Text))

Max Activating Token Index: 45

Text #18

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6645. Min Act: -0.0002

Data Index: 20310 (C4 (Web Text))

Max Activating Token Index: 15

Text #19

Max Range: 0.8420. Min Range: -0.8420

Max Act: 0.6847. Min Act: -0.0002

Data Index: 1388169 (Python Code)

Max Activating Token Index: 115