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

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

Neuron 774 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.9779. Min Act: -0.0003

Data Index: 913594 (C4 (Web Text))

Max Activating Token Index: 179

Text #1

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.7250. Min Act: -0.0003

Data Index: 134414 (C4 (Web Text))

Max Activating Token Index: 346

Text #2

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6647. Min Act: -0.0003

Data Index: 1162573 (C4 (Web Text))

Max Activating Token Index: 460

Text #3

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6693. Min Act: -0.0003

Data Index: 206163 (C4 (Web Text))

Max Activating Token Index: 317

Text #4

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6418. Min Act: -0.0003

Data Index: 195148 (C4 (Web Text))

Max Activating Token Index: 399

Text #5

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6338. Min Act: -0.0003

Data Index: 1224160 (C4 (Web Text))

Max Activating Token Index: 958

Text #6

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6344. Min Act: -0.0003

Data Index: 867739 (C4 (Web Text))

Max Activating Token Index: 560

Text #7

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6286. Min Act: -0.0003

Data Index: 153842 (C4 (Web Text))

Max Activating Token Index: 427

Text #8

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6245. Min Act: -0.0003

Data Index: 103661 (C4 (Web Text))

Max Activating Token Index: 458

Text #9

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6246. Min Act: -0.0003

Data Index: 32758 (C4 (Web Text))

Max Activating Token Index: 353

Text #10

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6117. Min Act: -0.0003

Data Index: 1299859 (C4 (Web Text))

Max Activating Token Index: 118

Text #11

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6104. Min Act: -0.0002

Data Index: 1292367 (C4 (Web Text))

Max Activating Token Index: 42

Text #12

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6066. Min Act: -0.0003

Data Index: 190301 (C4 (Web Text))

Max Activating Token Index: 844

Text #13

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.6307. Min Act: -0.0002

Data Index: 451182 (C4 (Web Text))

Max Activating Token Index: 64

Text #14

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.5942. Min Act: -0.0003

Data Index: 1236589 (C4 (Web Text))

Max Activating Token Index: 224

Text #15

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.5602. Min Act: -0.0003

Data Index: 156833 (C4 (Web Text))

Max Activating Token Index: 387

Text #16

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.5639. Min Act: -0.0003

Data Index: 1245334 (C4 (Web Text))

Max Activating Token Index: 504

Text #17

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.5616. Min Act: -0.0003

Data Index: 524028 (C4 (Web Text))

Max Activating Token Index: 364

Text #18

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.5609. Min Act: -0.0003

Data Index: 652935 (C4 (Web Text))

Max Activating Token Index: 834

Text #19

Max Range: 0.9779. Min Range: -0.9779

Max Act: 0.5588. Min Act: -0.0003

Data Index: 1113389 (C4 (Web Text))

Max Activating Token Index: 649