Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 6125 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.0355. Min Range: -3.0355

Max Act: 3.0355. Min Act: -0.0000

Data Index: 1150833 (C4 (Web Text))

Max Activating Token Index: 730

Text #1

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.8335. Min Act: -0.0000

Data Index: 1612948 (Python Code)

Max Activating Token Index: 76

Text #2

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.9700. Min Act: -0.0000

Data Index: 1328177 (C4 (Web Text))

Max Activating Token Index: 76

Text #3

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.7855. Min Act: -0.0000

Data Index: 1310615 (C4 (Web Text))

Max Activating Token Index: 727

Text #4

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.6494. Min Act: -0.0000

Data Index: 1488911 (Python Code)

Max Activating Token Index: 323

Text #5

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.6132. Min Act: -0.0000

Data Index: 375260 (C4 (Web Text))

Max Activating Token Index: 2

Text #6

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.4388. Min Act: -0.0000

Data Index: 1439575 (Python Code)

Max Activating Token Index: 13

Text #7

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.4279. Min Act: -0.0000

Data Index: 1065179 (C4 (Web Text))

Max Activating Token Index: 938

Text #8

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.4895. Min Act: -0.0000

Data Index: 1437526 (Python Code)

Max Activating Token Index: 370

Text #9

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.4827. Min Act: -0.0000

Data Index: 165517 (C4 (Web Text))

Max Activating Token Index: 915

Text #10

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.4335. Min Act: -0.0000

Data Index: 292284 (C4 (Web Text))

Max Activating Token Index: 849

Text #11

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.5072. Min Act: -0.0000

Data Index: 1122694 (C4 (Web Text))

Max Activating Token Index: 865

Text #12

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.3643. Min Act: -0.0000

Data Index: 1460746 (Python Code)

Max Activating Token Index: 125

Text #13

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.3826. Min Act: -0.0000

Data Index: 978338 (C4 (Web Text))

Max Activating Token Index: 905

Text #14

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.3734. Min Act: -0.0000

Data Index: 921100 (C4 (Web Text))

Max Activating Token Index: 140

Text #15

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.2081. Min Act: -0.0000

Data Index: 825977 (C4 (Web Text))

Max Activating Token Index: 18

Text #16

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.3164. Min Act: -0.0000

Data Index: 646591 (C4 (Web Text))

Max Activating Token Index: 734

Text #17

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.2907. Min Act: -0.0000

Data Index: 1379367 (Python Code)

Max Activating Token Index: 176

Text #18

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.3004. Min Act: -0.0000

Data Index: 1270370 (C4 (Web Text))

Max Activating Token Index: 94

Text #19

Max Range: 3.0355. Min Range: -3.0355

Max Act: 2.1491. Min Act: -0.0000

Data Index: 1256276 (C4 (Web Text))

Max Activating Token Index: 182