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

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

Neuron 4912 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.7704. Min Act: -0.0000

Data Index: 852621 (C4 (Web Text))

Max Activating Token Index: 570

Text #1

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.7164. Min Act: -0.0000

Data Index: 144362 (C4 (Web Text))

Max Activating Token Index: 184

Text #2

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.5665. Min Act: -0.0000

Data Index: 134792 (C4 (Web Text))

Max Activating Token Index: 794

Text #3

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.5329. Min Act: -0.0000

Data Index: 727418 (C4 (Web Text))

Max Activating Token Index: 136

Text #4

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.5849. Min Act: -0.0000

Data Index: 132402 (C4 (Web Text))

Max Activating Token Index: 518

Text #5

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.4559. Min Act: -0.0000

Data Index: 1316987 (C4 (Web Text))

Max Activating Token Index: 801

Text #6

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.4401. Min Act: -0.0000

Data Index: 1411205 (Python Code)

Max Activating Token Index: 1010

Text #7

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3559. Min Act: -0.0000

Data Index: 689921 (C4 (Web Text))

Max Activating Token Index: 189

Text #8

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3866. Min Act: -0.0000

Data Index: 1387629 (Python Code)

Max Activating Token Index: 942

Text #9

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.2867. Min Act: -0.0000

Data Index: 1646832 (Python Code)

Max Activating Token Index: 629

Text #10

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3494. Min Act: -0.0000

Data Index: 1289581 (C4 (Web Text))

Max Activating Token Index: 469

Text #11

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3850. Min Act: -0.0000

Data Index: 1498368 (Python Code)

Max Activating Token Index: 402

Text #12

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.4142. Min Act: -0.0000

Data Index: 1024987 (C4 (Web Text))

Max Activating Token Index: 544

Text #13

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3339. Min Act: -0.0000

Data Index: 1346095 (C4 (Web Text))

Max Activating Token Index: 102

Text #14

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3412. Min Act: -0.0000

Data Index: 1097895 (C4 (Web Text))

Max Activating Token Index: 854

Text #15

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3446. Min Act: -0.0000

Data Index: 1282938 (C4 (Web Text))

Max Activating Token Index: 46

Text #16

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.2528. Min Act: -0.0000

Data Index: 211104 (C4 (Web Text))

Max Activating Token Index: 422

Text #17

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.2133. Min Act: -0.0000

Data Index: 341724 (C4 (Web Text))

Max Activating Token Index: 711

Text #18

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.2126. Min Act: -0.0000

Data Index: 1157830 (C4 (Web Text))

Max Activating Token Index: 493

Text #19

Max Range: 1.7704. Min Range: -1.7704

Max Act: 1.3595. Min Act: -0.0000

Data Index: 347794 (C4 (Web Text))

Max Activating Token Index: 1019