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

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

Neuron 815 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: 0.0943. Min Range: -0.0943

Max Act: 0.0943. Min Act: -0.0002

Data Index: 1374550 (Python Code)

Max Activating Token Index: 690

Text #1

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0889. Min Act: -0.0002

Data Index: 278160 (C4 (Web Text))

Max Activating Token Index: 845

Text #2

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0827. Min Act: -0.0002

Data Index: 441821 (C4 (Web Text))

Max Activating Token Index: 504

Text #3

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0878. Min Act: -0.0002

Data Index: 1630876 (Python Code)

Max Activating Token Index: 90

Text #4

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0840. Min Act: -0.0002

Data Index: 1524324 (Python Code)

Max Activating Token Index: 179

Text #5

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0839. Min Act: -0.0002

Data Index: 1475599 (Python Code)

Max Activating Token Index: 591

Text #6

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0834. Min Act: -0.0001

Data Index: 1595810 (Python Code)

Max Activating Token Index: 393

Text #7

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0851. Min Act: -0.0002

Data Index: 1264517 (C4 (Web Text))

Max Activating Token Index: 581

Text #8

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0824. Min Act: -0.0002

Data Index: 35711 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0811. Min Act: -0.0002

Data Index: 1634648 (Python Code)

Max Activating Token Index: 506

Text #10

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0817. Min Act: -0.0002

Data Index: 708853 (C4 (Web Text))

Max Activating Token Index: 734

Text #11

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0823. Min Act: -0.0002

Data Index: 1232439 (C4 (Web Text))

Max Activating Token Index: 799

Text #12

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0789. Min Act: -0.0002

Data Index: 978602 (C4 (Web Text))

Max Activating Token Index: 404

Text #13

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0757. Min Act: -0.0002

Data Index: 990759 (C4 (Web Text))

Max Activating Token Index: 614

Text #14

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0797. Min Act: -0.0002

Data Index: 48236 (C4 (Web Text))

Max Activating Token Index: 1009

Text #15

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0783. Min Act: -0.0002

Data Index: 1594277 (Python Code)

Max Activating Token Index: 581

Text #16

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0801. Min Act: -0.0002

Data Index: 1059459 (C4 (Web Text))

Max Activating Token Index: 394

Text #17

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0783. Min Act: -0.0002

Data Index: 1376839 (Python Code)

Max Activating Token Index: 40

Text #18

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0777. Min Act: -0.0002

Data Index: 1639265 (Python Code)

Max Activating Token Index: 287

Text #19

Max Range: 0.0943. Min Range: -0.0943

Max Act: 0.0761. Min Act: -0.0002

Data Index: 1113841 (C4 (Web Text))

Max Activating Token Index: 222