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

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

Neuron 1548 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.1057. Min Act: -0.0001

Data Index: 1485473 (Python Code)

Max Activating Token Index: 456

Text #1

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0872. Min Act: -0.0001

Data Index: 590927 (C4 (Web Text))

Max Activating Token Index: 190

Text #2

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0787. Min Act: -0.0001

Data Index: 1316026 (C4 (Web Text))

Max Activating Token Index: 40

Text #3

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0803. Min Act: -0.0001

Data Index: 41673 (C4 (Web Text))

Max Activating Token Index: 423

Text #4

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0784. Min Act: -0.0001

Data Index: 1124007 (C4 (Web Text))

Max Activating Token Index: 681

Text #5

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0751. Min Act: -0.0001

Data Index: 305365 (C4 (Web Text))

Max Activating Token Index: 599

Text #6

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0770. Min Act: -0.0001

Data Index: 973266 (C4 (Web Text))

Max Activating Token Index: 279

Text #7

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0756. Min Act: -0.0001

Data Index: 1472367 (Python Code)

Max Activating Token Index: 534

Text #8

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0780. Min Act: -0.0001

Data Index: 1597181 (Python Code)

Max Activating Token Index: 726

Text #9

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0715. Min Act: -0.0001

Data Index: 1300477 (C4 (Web Text))

Max Activating Token Index: 31

Text #10

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0728. Min Act: -0.0001

Data Index: 733432 (C4 (Web Text))

Max Activating Token Index: 551

Text #11

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0695. Min Act: -0.0001

Data Index: 1493415 (Python Code)

Max Activating Token Index: 55

Text #12

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0738. Min Act: -0.0001

Data Index: 1409779 (Python Code)

Max Activating Token Index: 650

Text #13

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0702. Min Act: -0.0001

Data Index: 1450232 (Python Code)

Max Activating Token Index: 154

Text #14

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0668. Min Act: -0.0001

Data Index: 1629313 (Python Code)

Max Activating Token Index: 65

Text #15

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0651. Min Act: -0.0001

Data Index: 1387813 (Python Code)

Max Activating Token Index: 336

Text #16

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0686. Min Act: -0.0001

Data Index: 354326 (C4 (Web Text))

Max Activating Token Index: 364

Text #17

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0658. Min Act: -0.0001

Data Index: 839953 (C4 (Web Text))

Max Activating Token Index: 725

Text #18

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0617. Min Act: -0.0001

Data Index: 819724 (C4 (Web Text))

Max Activating Token Index: 10

Text #19

Max Range: 0.1057. Min Range: -0.1057

Max Act: 0.0622. Min Act: -0.0001

Data Index: 1413606 (Python Code)

Max Activating Token Index: 563