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

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

Neuron 64 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6824. Min Act: -0.0001

Data Index: 849483 (C4 (Web Text))

Max Activating Token Index: 936

Text #1

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6578. Min Act: -0.0001

Data Index: 922617 (C4 (Web Text))

Max Activating Token Index: 95

Text #2

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6231. Min Act: -0.0001

Data Index: 1215982 (C4 (Web Text))

Max Activating Token Index: 556

Text #3

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6530. Min Act: -0.0001

Data Index: 481817 (C4 (Web Text))

Max Activating Token Index: 262

Text #4

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6265. Min Act: -0.0001

Data Index: 1167857 (C4 (Web Text))

Max Activating Token Index: 68

Text #5

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6236. Min Act: -0.0001

Data Index: 609564 (C4 (Web Text))

Max Activating Token Index: 236

Text #6

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6224. Min Act: -0.0001

Data Index: 950412 (C4 (Web Text))

Max Activating Token Index: 43

Text #7

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6188. Min Act: -0.0001

Data Index: 934332 (C4 (Web Text))

Max Activating Token Index: 42

Text #8

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.6192. Min Act: -0.0001

Data Index: 535286 (C4 (Web Text))

Max Activating Token Index: 229

Text #9

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5917. Min Act: -0.0001

Data Index: 1336813 (C4 (Web Text))

Max Activating Token Index: 428

Text #10

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5965. Min Act: -0.0001

Data Index: 435829 (C4 (Web Text))

Max Activating Token Index: 783

Text #11

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5766. Min Act: -0.0001

Data Index: 160219 (C4 (Web Text))

Max Activating Token Index: 47

Text #12

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5852. Min Act: -0.0001

Data Index: 100109 (C4 (Web Text))

Max Activating Token Index: 294

Text #13

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5643. Min Act: -0.0001

Data Index: 541972 (C4 (Web Text))

Max Activating Token Index: 55

Text #14

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5825. Min Act: -0.0001

Data Index: 287640 (C4 (Web Text))

Max Activating Token Index: 69

Text #15

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5806. Min Act: -0.0001

Data Index: 1148185 (C4 (Web Text))

Max Activating Token Index: 267

Text #16

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5838. Min Act: -0.0001

Data Index: 545516 (C4 (Web Text))

Max Activating Token Index: 928

Text #17

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5611. Min Act: -0.0001

Data Index: 374706 (C4 (Web Text))

Max Activating Token Index: 133

Text #18

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5756. Min Act: -0.0001

Data Index: 981270 (C4 (Web Text))

Max Activating Token Index: 432

Text #19

Max Range: 0.6824. Min Range: -0.6824

Max Act: 0.5596. Min Act: -0.0001

Data Index: 956754 (C4 (Web Text))

Max Activating Token Index: 512