Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2069 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.6869. Min Act: -0.0003

Data Index: 78273 (C4 (Web Text))

Max Activating Token Index: 660

Text #1

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.5396. Min Act: -0.0003

Data Index: 927624 (C4 (Web Text))

Max Activating Token Index: 592

Text #2

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4525. Min Act: -0.0003

Data Index: 1098951 (C4 (Web Text))

Max Activating Token Index: 616

Text #3

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.5751. Min Act: -0.0003

Data Index: 6006 (C4 (Web Text))

Max Activating Token Index: 278

Text #4

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.5041. Min Act: -0.0003

Data Index: 170819 (C4 (Web Text))

Max Activating Token Index: 600

Text #5

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4631. Min Act: -0.0003

Data Index: 147105 (C4 (Web Text))

Max Activating Token Index: 78

Text #6

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4143. Min Act: -0.0003

Data Index: 1272329 (C4 (Web Text))

Max Activating Token Index: 597

Text #7

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4427. Min Act: -0.0003

Data Index: 915746 (C4 (Web Text))

Max Activating Token Index: 29

Text #8

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4268. Min Act: -0.0003

Data Index: 931901 (C4 (Web Text))

Max Activating Token Index: 920

Text #9

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4622. Min Act: -0.0003

Data Index: 1076025 (C4 (Web Text))

Max Activating Token Index: 900

Text #10

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4184. Min Act: -0.0003

Data Index: 54121 (C4 (Web Text))

Max Activating Token Index: 210

Text #11

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4299. Min Act: -0.0003

Data Index: 1592460 (Python Code)

Max Activating Token Index: 86

Text #12

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4311. Min Act: -0.0003

Data Index: 1650410 (Python Code)

Max Activating Token Index: 123

Text #13

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.3931. Min Act: -0.0003

Data Index: 818573 (C4 (Web Text))

Max Activating Token Index: 199

Text #14

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4449. Min Act: -0.0003

Data Index: 1548474 (Python Code)

Max Activating Token Index: 422

Text #15

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4090. Min Act: -0.0003

Data Index: 476148 (C4 (Web Text))

Max Activating Token Index: 375

Text #16

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4236. Min Act: -0.0003

Data Index: 1597373 (Python Code)

Max Activating Token Index: 143

Text #17

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4340. Min Act: -0.0003

Data Index: 902169 (C4 (Web Text))

Max Activating Token Index: 42

Text #18

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4126. Min Act: -0.0003

Data Index: 1520836 (Python Code)

Max Activating Token Index: 195

Text #19

Max Range: 0.6869. Min Range: -0.6869

Max Act: 0.4019. Min Act: -0.0003

Data Index: 474648 (C4 (Web Text))

Max Activating Token Index: 2