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

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

Neuron 197 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.3842. Min Range: -4.3842

Max Act: 4.3369. Min Act: -0.0000

Data Index: 76562 (C4 (Web Text))

Max Activating Token Index: 845

Text #1

Max Range: 4.3842. Min Range: -4.3842

Max Act: 4.3842. Min Act: -0.0000

Data Index: 854218 (C4 (Web Text))

Max Activating Token Index: 680

Text #2

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.6484. Min Act: -0.0000

Data Index: 1613528 (Python Code)

Max Activating Token Index: 477

Text #3

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.3675. Min Act: -0.0000

Data Index: 1032424 (C4 (Web Text))

Max Activating Token Index: 105

Text #4

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.4965. Min Act: -0.0000

Data Index: 160643 (C4 (Web Text))

Max Activating Token Index: 281

Text #5

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.5917. Min Act: -0.0000

Data Index: 299519 (C4 (Web Text))

Max Activating Token Index: 340

Text #6

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.1049. Min Act: -0.0000

Data Index: 40504 (C4 (Web Text))

Max Activating Token Index: 96

Text #7

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.9402. Min Act: -0.0000

Data Index: 1645048 (Python Code)

Max Activating Token Index: 516

Text #8

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.2372. Min Act: -0.0000

Data Index: 1272170 (C4 (Web Text))

Max Activating Token Index: 446

Text #9

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.2679. Min Act: -0.0000

Data Index: 281462 (C4 (Web Text))

Max Activating Token Index: 271

Text #10

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.5301. Min Act: -0.0000

Data Index: 750435 (C4 (Web Text))

Max Activating Token Index: 269

Text #11

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.0455. Min Act: -0.0000

Data Index: 251658 (C4 (Web Text))

Max Activating Token Index: 29

Text #12

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.8929. Min Act: -0.0000

Data Index: 787149 (C4 (Web Text))

Max Activating Token Index: 413

Text #13

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.9806. Min Act: -0.0000

Data Index: 1462915 (Python Code)

Max Activating Token Index: 126

Text #14

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.9750. Min Act: -0.0000

Data Index: 1485015 (Python Code)

Max Activating Token Index: 41

Text #15

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.8250. Min Act: -0.0000

Data Index: 1232027 (C4 (Web Text))

Max Activating Token Index: 84

Text #16

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.2282. Min Act: -0.0000

Data Index: 976682 (C4 (Web Text))

Max Activating Token Index: 997

Text #17

Max Range: 4.3842. Min Range: -4.3842

Max Act: 3.0727. Min Act: -0.0000

Data Index: 798798 (C4 (Web Text))

Max Activating Token Index: 758

Text #18

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.9264. Min Act: -0.0000

Data Index: 885301 (C4 (Web Text))

Max Activating Token Index: 346

Text #19

Max Range: 4.3842. Min Range: -4.3842

Max Act: 2.7430. Min Act: -0.0000

Data Index: 380285 (C4 (Web Text))

Max Activating Token Index: 29