Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 775 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.1630. Min Act: -0.0002

Data Index: 872525 (C4 (Web Text))

Max Activating Token Index: 841

Text #1

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0912. Min Act: -0.0002

Data Index: 1187822 (C4 (Web Text))

Max Activating Token Index: 203

Text #2

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0660. Min Act: -0.0002

Data Index: 685042 (C4 (Web Text))

Max Activating Token Index: 771

Text #3

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0607. Min Act: -0.0002

Data Index: 192909 (C4 (Web Text))

Max Activating Token Index: 921

Text #4

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0588. Min Act: -0.0002

Data Index: 1469967 (Python Code)

Max Activating Token Index: 927

Text #5

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0315. Min Act: -0.0002

Data Index: 61073 (C4 (Web Text))

Max Activating Token Index: 736

Text #6

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0552. Min Act: -0.0002

Data Index: 1457232 (Python Code)

Max Activating Token Index: 527

Text #7

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0501. Min Act: -0.0002

Data Index: 458562 (C4 (Web Text))

Max Activating Token Index: 182

Text #8

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0209. Min Act: -0.0002

Data Index: 1557033 (Python Code)

Max Activating Token Index: 980

Text #9

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0215. Min Act: -0.0002

Data Index: 29955 (C4 (Web Text))

Max Activating Token Index: 1000

Text #10

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0503. Min Act: -0.0002

Data Index: 1057684 (C4 (Web Text))

Max Activating Token Index: 410

Text #11

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0521. Min Act: -0.0002

Data Index: 1459077 (Python Code)

Max Activating Token Index: 852

Text #12

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0499. Min Act: -0.0002

Data Index: 1620333 (Python Code)

Max Activating Token Index: 649

Text #13

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0506. Min Act: -0.0002

Data Index: 1608103 (Python Code)

Max Activating Token Index: 188

Text #14

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0448. Min Act: -0.0002

Data Index: 323549 (C4 (Web Text))

Max Activating Token Index: 753

Text #15

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0465. Min Act: -0.0002

Data Index: 609725 (C4 (Web Text))

Max Activating Token Index: 829

Text #16

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0475. Min Act: -0.0002

Data Index: 1523338 (Python Code)

Max Activating Token Index: 618

Text #17

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0470. Min Act: -0.0002

Data Index: 1360250 (Python Code)

Max Activating Token Index: 758

Text #18

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0460. Min Act: -0.0002

Data Index: 77647 (C4 (Web Text))

Max Activating Token Index: 428

Text #19

Max Range: 0.1630. Min Range: -0.1630

Max Act: 0.0460. Min Act: -0.0002

Data Index: 1535301 (Python Code)

Max Activating Token Index: 985