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

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

Neuron 1890 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.6237. Min Act: -0.0000

Data Index: 296507 (C4 (Web Text))

Max Activating Token Index: 91

Text #1

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.5533. Min Act: -0.0000

Data Index: 748444 (C4 (Web Text))

Max Activating Token Index: 327

Text #2

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.3732. Min Act: -0.0000

Data Index: 582888 (C4 (Web Text))

Max Activating Token Index: 709

Text #3

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.5136. Min Act: -0.0000

Data Index: 455015 (C4 (Web Text))

Max Activating Token Index: 142

Text #4

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.3140. Min Act: -0.0000

Data Index: 804202 (C4 (Web Text))

Max Activating Token Index: 205

Text #5

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1678. Min Act: -0.0000

Data Index: 1205787 (C4 (Web Text))

Max Activating Token Index: 34

Text #6

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1854. Min Act: -0.0000

Data Index: 306797 (C4 (Web Text))

Max Activating Token Index: 482

Text #7

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1838. Min Act: -0.0000

Data Index: 57381 (C4 (Web Text))

Max Activating Token Index: 47

Text #8

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1253. Min Act: -0.0000

Data Index: 1552015 (Python Code)

Max Activating Token Index: 880

Text #9

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1057. Min Act: -0.0000

Data Index: 1503006 (Python Code)

Max Activating Token Index: 727

Text #10

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1028. Min Act: -0.0000

Data Index: 1067557 (C4 (Web Text))

Max Activating Token Index: 572

Text #11

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0836. Min Act: -0.0000

Data Index: 1468886 (Python Code)

Max Activating Token Index: 643

Text #12

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0825. Min Act: -0.0000

Data Index: 1437573 (Python Code)

Max Activating Token Index: 57

Text #13

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0714. Min Act: -0.0000

Data Index: 1494975 (Python Code)

Max Activating Token Index: 155

Text #14

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.1015. Min Act: -0.0000

Data Index: 1447148 (Python Code)

Max Activating Token Index: 648

Text #15

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0580. Min Act: -0.0000

Data Index: 756749 (C4 (Web Text))

Max Activating Token Index: 399

Text #16

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0422. Min Act: -0.0000

Data Index: 1628593 (Python Code)

Max Activating Token Index: 990

Text #17

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0527. Min Act: -0.0000

Data Index: 1462028 (Python Code)

Max Activating Token Index: 806

Text #18

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0124. Min Act: -0.0000

Data Index: 1481640 (Python Code)

Max Activating Token Index: 564

Text #19

Max Range: 1.6237. Min Range: -1.6237

Max Act: 1.0781. Min Act: -0.0000

Data Index: 1467742 (Python Code)

Max Activating Token Index: 544