Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2604 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.2162. Min Act: -0.0001

Data Index: 270460 (C4 (Web Text))

Max Activating Token Index: 15

Text #1

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.2049. Min Act: -0.0001

Data Index: 141515 (C4 (Web Text))

Max Activating Token Index: 967

Text #2

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1862. Min Act: -0.0002

Data Index: 1281897 (C4 (Web Text))

Max Activating Token Index: 936

Text #3

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1818. Min Act: -0.0002

Data Index: 1075556 (C4 (Web Text))

Max Activating Token Index: 438

Text #4

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1755. Min Act: -0.0002

Data Index: 351033 (C4 (Web Text))

Max Activating Token Index: 975

Text #5

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1745. Min Act: -0.0002

Data Index: 1346454 (C4 (Web Text))

Max Activating Token Index: 315

Text #6

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1736. Min Act: -0.0002

Data Index: 384042 (C4 (Web Text))

Max Activating Token Index: 275

Text #7

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1600. Min Act: -0.0002

Data Index: 1058967 (C4 (Web Text))

Max Activating Token Index: 33

Text #8

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1731. Min Act: -0.0002

Data Index: 840690 (C4 (Web Text))

Max Activating Token Index: 882

Text #9

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1587. Min Act: -0.0002

Data Index: 185727 (C4 (Web Text))

Max Activating Token Index: 180

Text #10

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1629. Min Act: -0.0001

Data Index: 846202 (C4 (Web Text))

Max Activating Token Index: 43

Text #11

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1495. Min Act: -0.0001

Data Index: 1613457 (Python Code)

Max Activating Token Index: 973

Text #12

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1622. Min Act: -0.0001

Data Index: 109576 (C4 (Web Text))

Max Activating Token Index: 109

Text #13

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1750. Min Act: -0.0001

Data Index: 835841 (C4 (Web Text))

Max Activating Token Index: 219

Text #14

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1675. Min Act: -0.0001

Data Index: 875073 (C4 (Web Text))

Max Activating Token Index: 1004

Text #15

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1678. Min Act: -0.0002

Data Index: 942579 (C4 (Web Text))

Max Activating Token Index: 953

Text #16

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1698. Min Act: -0.0001

Data Index: 1087246 (C4 (Web Text))

Max Activating Token Index: 467

Text #17

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1587. Min Act: -0.0001

Data Index: 17685 (C4 (Web Text))

Max Activating Token Index: 528

Text #18

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1712. Min Act: -0.0002

Data Index: 496670 (C4 (Web Text))

Max Activating Token Index: 565

Text #19

Max Range: 0.2162. Min Range: -0.2162

Max Act: 0.1635. Min Act: -0.0002

Data Index: 75532 (C4 (Web Text))

Max Activating Token Index: 93