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

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

Neuron 4583 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: 2.4285. Min Range: -2.4285

Max Act: 2.3353. Min Act: -0.0000

Data Index: 1239450 (C4 (Web Text))

Max Activating Token Index: 919

Text #1

Max Range: 2.4285. Min Range: -2.4285

Max Act: 2.4285. Min Act: -0.0000

Data Index: 1216590 (C4 (Web Text))

Max Activating Token Index: 194

Text #2

Max Range: 2.4285. Min Range: -2.4285

Max Act: 2.1676. Min Act: -0.0000

Data Index: 830514 (C4 (Web Text))

Max Activating Token Index: 475

Text #3

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.9876. Min Act: -0.0000

Data Index: 914172 (C4 (Web Text))

Max Activating Token Index: 106

Text #4

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.7546. Min Act: -0.0000

Data Index: 810267 (C4 (Web Text))

Max Activating Token Index: 336

Text #5

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.8222. Min Act: -0.0000

Data Index: 950576 (C4 (Web Text))

Max Activating Token Index: 92

Text #6

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.5634. Min Act: -0.0000

Data Index: 1167814 (C4 (Web Text))

Max Activating Token Index: 337

Text #7

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.8638. Min Act: -0.0000

Data Index: 1276114 (C4 (Web Text))

Max Activating Token Index: 820

Text #8

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.7210. Min Act: -0.0000

Data Index: 869412 (C4 (Web Text))

Max Activating Token Index: 273

Text #9

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.7742. Min Act: -0.0000

Data Index: 1612577 (Python Code)

Max Activating Token Index: 13

Text #10

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.6726. Min Act: -0.0000

Data Index: 1024383 (C4 (Web Text))

Max Activating Token Index: 634

Text #11

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.5241. Min Act: -0.0000

Data Index: 1062163 (C4 (Web Text))

Max Activating Token Index: 716

Text #12

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.6498. Min Act: -0.0000

Data Index: 482137 (C4 (Web Text))

Max Activating Token Index: 962

Text #13

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.4362. Min Act: -0.0000

Data Index: 810879 (C4 (Web Text))

Max Activating Token Index: 84

Text #14

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.6062. Min Act: -0.0000

Data Index: 1361248 (Python Code)

Max Activating Token Index: 48

Text #15

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.5737. Min Act: -0.0000

Data Index: 497953 (C4 (Web Text))

Max Activating Token Index: 195

Text #16

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.4880. Min Act: -0.0000

Data Index: 1615056 (Python Code)

Max Activating Token Index: 62

Text #17

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.3738. Min Act: -0.0000

Data Index: 1047634 (C4 (Web Text))

Max Activating Token Index: 168

Text #18

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.5708. Min Act: -0.0000

Data Index: 614707 (C4 (Web Text))

Max Activating Token Index: 646

Text #19

Max Range: 2.4285. Min Range: -2.4285

Max Act: 1.6511. Min Act: -0.0000

Data Index: 1161709 (C4 (Web Text))

Max Activating Token Index: 897