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

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

Neuron 3070 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0900. Min Act: -0.0002

Data Index: 1609252 (Python Code)

Max Activating Token Index: 22

Text #1

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0785. Min Act: -0.0002

Data Index: 1561311 (Python Code)

Max Activating Token Index: 280

Text #2

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0786. Min Act: -0.0002

Data Index: 697237 (C4 (Web Text))

Max Activating Token Index: 283

Text #3

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0754. Min Act: -0.0002

Data Index: 231429 (C4 (Web Text))

Max Activating Token Index: 564

Text #4

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0745. Min Act: -0.0002

Data Index: 1107410 (C4 (Web Text))

Max Activating Token Index: 409

Text #5

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0697. Min Act: -0.0002

Data Index: 583452 (C4 (Web Text))

Max Activating Token Index: 434

Text #6

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0711. Min Act: -0.0002

Data Index: 1448658 (Python Code)

Max Activating Token Index: 435

Text #7

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0697. Min Act: -0.0002

Data Index: 1461579 (Python Code)

Max Activating Token Index: 688

Text #8

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0693. Min Act: -0.0002

Data Index: 1633221 (Python Code)

Max Activating Token Index: 482

Text #9

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0661. Min Act: -0.0002

Data Index: 556344 (C4 (Web Text))

Max Activating Token Index: 529

Text #10

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0649. Min Act: -0.0002

Data Index: 1469598 (Python Code)

Max Activating Token Index: 60

Text #11

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0637. Min Act: -0.0002

Data Index: 652632 (C4 (Web Text))

Max Activating Token Index: 365

Text #12

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0628. Min Act: -0.0002

Data Index: 1360771 (Python Code)

Max Activating Token Index: 218

Text #13

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0620. Min Act: -0.0002

Data Index: 1436611 (Python Code)

Max Activating Token Index: 29

Text #14

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0650. Min Act: -0.0002

Data Index: 1427949 (Python Code)

Max Activating Token Index: 187

Text #15

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0598. Min Act: -0.0002

Data Index: 1609208 (Python Code)

Max Activating Token Index: 577

Text #16

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0608. Min Act: -0.0002

Data Index: 132318 (C4 (Web Text))

Max Activating Token Index: 980

Text #17

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0637. Min Act: -0.0002

Data Index: 270279 (C4 (Web Text))

Max Activating Token Index: 722

Text #18

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0642. Min Act: -0.0002

Data Index: 272541 (C4 (Web Text))

Max Activating Token Index: 123

Text #19

Max Range: 0.0900. Min Range: -0.0900

Max Act: 0.0606. Min Act: -0.0002

Data Index: 1432579 (Python Code)

Max Activating Token Index: 369