Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1809 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.1603. Min Act: -0.0006

Data Index: 1504280 (Python Code)

Max Activating Token Index: 102

Text #1

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.0778. Min Act: -0.0006

Data Index: 994913 (C4 (Web Text))

Max Activating Token Index: 408

Text #2

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.1239. Min Act: -0.0006

Data Index: 985788 (C4 (Web Text))

Max Activating Token Index: 361

Text #3

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.0188. Min Act: -0.0006

Data Index: 1098961 (C4 (Web Text))

Max Activating Token Index: 29

Text #4

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.1514. Min Act: -0.0006

Data Index: 1385715 (Python Code)

Max Activating Token Index: 205

Text #5

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.1411. Min Act: -0.0006

Data Index: 1507328 (Python Code)

Max Activating Token Index: 914

Text #6

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.9595. Min Act: -0.0006

Data Index: 749772 (C4 (Web Text))

Max Activating Token Index: 234

Text #7

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.6682. Min Act: -0.0006

Data Index: 686494 (C4 (Web Text))

Max Activating Token Index: 25

Text #8

Max Range: 3.1603. Min Range: -3.1603

Max Act: 3.0615. Min Act: -0.0006

Data Index: 1509880 (Python Code)

Max Activating Token Index: 515

Text #9

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.9038. Min Act: -0.0006

Data Index: 108925 (C4 (Web Text))

Max Activating Token Index: 371

Text #10

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.8780. Min Act: -0.0006

Data Index: 1526546 (Python Code)

Max Activating Token Index: 336

Text #11

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.7893. Min Act: -0.0006

Data Index: 426015 (C4 (Web Text))

Max Activating Token Index: 940

Text #12

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.6933. Min Act: -0.0006

Data Index: 1510593 (Python Code)

Max Activating Token Index: 63

Text #13

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.6620. Min Act: -0.0006

Data Index: 164947 (C4 (Web Text))

Max Activating Token Index: 744

Text #14

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.6863. Min Act: -0.0006

Data Index: 1364252 (Python Code)

Max Activating Token Index: 75

Text #15

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.5739. Min Act: -0.0006

Data Index: 1445175 (Python Code)

Max Activating Token Index: 6

Text #16

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.6298. Min Act: -0.0006

Data Index: 141330 (C4 (Web Text))

Max Activating Token Index: 596

Text #17

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.3529. Min Act: -0.0006

Data Index: 789665 (C4 (Web Text))

Max Activating Token Index: 914

Text #18

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.4838. Min Act: -0.0006

Data Index: 1076712 (C4 (Web Text))

Max Activating Token Index: 952

Text #19

Max Range: 3.1603. Min Range: -3.1603

Max Act: 2.4804. Min Act: -0.0006

Data Index: 1650626 (Python Code)

Max Activating Token Index: 834