Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3402 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9879. Min Act: -0.0002

Data Index: 124406 (C4 (Web Text))

Max Activating Token Index: 930

Text #1

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0512. Min Act: -0.0002

Data Index: 1087671 (C4 (Web Text))

Max Activating Token Index: 522

Text #2

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0656. Min Act: -0.0002

Data Index: 1332715 (C4 (Web Text))

Max Activating Token Index: 459

Text #3

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0955. Min Act: -0.0002

Data Index: 282983 (C4 (Web Text))

Max Activating Token Index: 73

Text #4

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0622. Min Act: -0.0002

Data Index: 177894 (C4 (Web Text))

Max Activating Token Index: 71

Text #5

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0562. Min Act: -0.0002

Data Index: 1028748 (C4 (Web Text))

Max Activating Token Index: 85

Text #6

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0050. Min Act: -0.0002

Data Index: 1310128 (C4 (Web Text))

Max Activating Token Index: 123

Text #7

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9664. Min Act: -0.0002

Data Index: 1504351 (Python Code)

Max Activating Token Index: 605

Text #8

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0207. Min Act: -0.0002

Data Index: 583958 (C4 (Web Text))

Max Activating Token Index: 206

Text #9

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9508. Min Act: -0.0002

Data Index: 103038 (C4 (Web Text))

Max Activating Token Index: 657

Text #10

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9915. Min Act: -0.0002

Data Index: 1077169 (C4 (Web Text))

Max Activating Token Index: 900

Text #11

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0530. Min Act: -0.0002

Data Index: 1038838 (C4 (Web Text))

Max Activating Token Index: 790

Text #12

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.8996. Min Act: -0.0002

Data Index: 957960 (C4 (Web Text))

Max Activating Token Index: 152

Text #13

Max Range: 1.0955. Min Range: -1.0955

Max Act: 1.0063. Min Act: -0.0002

Data Index: 885590 (C4 (Web Text))

Max Activating Token Index: 948

Text #14

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9667. Min Act: -0.0002

Data Index: 412018 (C4 (Web Text))

Max Activating Token Index: 498

Text #15

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9603. Min Act: -0.0002

Data Index: 736602 (C4 (Web Text))

Max Activating Token Index: 38

Text #16

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9490. Min Act: -0.0002

Data Index: 831728 (C4 (Web Text))

Max Activating Token Index: 625

Text #17

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.7960. Min Act: -0.0002

Data Index: 35658 (C4 (Web Text))

Max Activating Token Index: 236

Text #18

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.8769. Min Act: -0.0002

Data Index: 1550720 (Python Code)

Max Activating Token Index: 352

Text #19

Max Range: 1.0955. Min Range: -1.0955

Max Act: 0.9563. Min Act: -0.0002

Data Index: 1627981 (Python Code)

Max Activating Token Index: 425