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

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

Neuron 3003 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4717. Min Range: -1.4717

Max Act: 1.3873. Min Act: -0.0000

Data Index: 815597 (C4 (Web Text))

Max Activating Token Index: 50

Text #1

Max Range: 1.4717. Min Range: -1.4717

Max Act: 1.4717. Min Act: -0.0000

Data Index: 387759 (C4 (Web Text))

Max Activating Token Index: 390

Text #2

Max Range: 1.4717. Min Range: -1.4717

Max Act: 1.0973. Min Act: -0.0000

Data Index: 1300513 (C4 (Web Text))

Max Activating Token Index: 690

Text #3

Max Range: 1.4717. Min Range: -1.4717

Max Act: 1.1954. Min Act: -0.0000

Data Index: 1468716 (Python Code)

Max Activating Token Index: 443

Text #4

Max Range: 1.4717. Min Range: -1.4717

Max Act: 1.1153. Min Act: -0.0000

Data Index: 790976 (C4 (Web Text))

Max Activating Token Index: 18

Text #5

Max Range: 1.4717. Min Range: -1.4717

Max Act: 1.0463. Min Act: -0.0000

Data Index: 1441136 (Python Code)

Max Activating Token Index: 315

Text #6

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.9180. Min Act: -0.0000

Data Index: 1524875 (Python Code)

Max Activating Token Index: 827

Text #7

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.9217. Min Act: -0.0000

Data Index: 469916 (C4 (Web Text))

Max Activating Token Index: 836

Text #8

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.9636. Min Act: -0.0000

Data Index: 1237132 (C4 (Web Text))

Max Activating Token Index: 746

Text #9

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8563. Min Act: -0.0000

Data Index: 1293904 (C4 (Web Text))

Max Activating Token Index: 254

Text #10

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8210. Min Act: -0.0000

Data Index: 151153 (C4 (Web Text))

Max Activating Token Index: 749

Text #11

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8443. Min Act: -0.0000

Data Index: 998077 (C4 (Web Text))

Max Activating Token Index: 777

Text #12

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8600. Min Act: -0.0000

Data Index: 1086306 (C4 (Web Text))

Max Activating Token Index: 523

Text #13

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8051. Min Act: -0.0000

Data Index: 1654637 (Python Code)

Max Activating Token Index: 749

Text #14

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8705. Min Act: -0.0000

Data Index: 779117 (C4 (Web Text))

Max Activating Token Index: 271

Text #15

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.7871. Min Act: -0.0000

Data Index: 1181172 (C4 (Web Text))

Max Activating Token Index: 592

Text #16

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.7819. Min Act: -0.0000

Data Index: 760475 (C4 (Web Text))

Max Activating Token Index: 309

Text #17

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.7461. Min Act: -0.0000

Data Index: 385137 (C4 (Web Text))

Max Activating Token Index: 606

Text #18

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8807. Min Act: -0.0000

Data Index: 1038344 (C4 (Web Text))

Max Activating Token Index: 752

Text #19

Max Range: 1.4717. Min Range: -1.4717

Max Act: 0.8733. Min Act: -0.0000

Data Index: 862384 (C4 (Web Text))

Max Activating Token Index: 601