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

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

Neuron 2459 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.1505. Min Act: -0.0001

Data Index: 715307 (C4 (Web Text))

Max Activating Token Index: 990

Text #1

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.1394. Min Act: -0.0001

Data Index: 1050924 (C4 (Web Text))

Max Activating Token Index: 313

Text #2

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.1019. Min Act: -0.0001

Data Index: 1005434 (C4 (Web Text))

Max Activating Token Index: 643

Text #3

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0896. Min Act: -0.0001

Data Index: 1351118 (C4 (Web Text))

Max Activating Token Index: 73

Text #4

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0595. Min Act: -0.0001

Data Index: 1169322 (C4 (Web Text))

Max Activating Token Index: 962

Text #5

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.1640. Min Act: -0.0001

Data Index: 527855 (C4 (Web Text))

Max Activating Token Index: 207

Text #6

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0183. Min Act: -0.0001

Data Index: 672048 (C4 (Web Text))

Max Activating Token Index: 43

Text #7

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0343. Min Act: -0.0001

Data Index: 120414 (C4 (Web Text))

Max Activating Token Index: 472

Text #8

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0159. Min Act: -0.0001

Data Index: 84847 (C4 (Web Text))

Max Activating Token Index: 674

Text #9

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9378. Min Act: -0.0001

Data Index: 769619 (C4 (Web Text))

Max Activating Token Index: 620

Text #10

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9714. Min Act: -0.0001

Data Index: 1110801 (C4 (Web Text))

Max Activating Token Index: 715

Text #11

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0575. Min Act: -0.0001

Data Index: 1473428 (Python Code)

Max Activating Token Index: 689

Text #12

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9518. Min Act: -0.0001

Data Index: 1196781 (C4 (Web Text))

Max Activating Token Index: 422

Text #13

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9691. Min Act: -0.0001

Data Index: 702181 (C4 (Web Text))

Max Activating Token Index: 878

Text #14

Max Range: 1.1640. Min Range: -1.1640

Max Act: 1.0036. Min Act: -0.0001

Data Index: 1187635 (C4 (Web Text))

Max Activating Token Index: 120

Text #15

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9688. Min Act: -0.0001

Data Index: 368449 (C4 (Web Text))

Max Activating Token Index: 605

Text #16

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9390. Min Act: -0.0001

Data Index: 537227 (C4 (Web Text))

Max Activating Token Index: 375

Text #17

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9388. Min Act: -0.0001

Data Index: 1554149 (Python Code)

Max Activating Token Index: 424

Text #18

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9316. Min Act: -0.0001

Data Index: 905565 (C4 (Web Text))

Max Activating Token Index: 196

Text #19

Max Range: 1.1640. Min Range: -1.1640

Max Act: 0.9708. Min Act: -0.0001

Data Index: 689418 (C4 (Web Text))

Max Activating Token Index: 393