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

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

Neuron 545 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: 3.6587. Min Range: -3.6587

Max Act: 3.6587. Min Act: -0.0000

Data Index: 243666 (C4 (Web Text))

Max Activating Token Index: 865

Text #1

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.9017. Min Act: -0.0000

Data Index: 262093 (C4 (Web Text))

Max Activating Token Index: 15

Text #2

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.8292. Min Act: -0.0000

Data Index: 360563 (C4 (Web Text))

Max Activating Token Index: 43

Text #3

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.9195. Min Act: -0.0000

Data Index: 1169315 (C4 (Web Text))

Max Activating Token Index: 265

Text #4

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.8656. Min Act: -0.0000

Data Index: 1641663 (Python Code)

Max Activating Token Index: 722

Text #5

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.7797. Min Act: -0.0000

Data Index: 908782 (C4 (Web Text))

Max Activating Token Index: 156

Text #6

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.9254. Min Act: -0.0000

Data Index: 932656 (C4 (Web Text))

Max Activating Token Index: 765

Text #7

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.7597. Min Act: -0.0000

Data Index: 2676 (C4 (Web Text))

Max Activating Token Index: 454

Text #8

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.7619. Min Act: -0.0000

Data Index: 1405454 (Python Code)

Max Activating Token Index: 254

Text #9

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.7585. Min Act: -0.0000

Data Index: 1123465 (C4 (Web Text))

Max Activating Token Index: 626

Text #10

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.8522. Min Act: -0.0000

Data Index: 601682 (C4 (Web Text))

Max Activating Token Index: 677

Text #11

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.6658. Min Act: -0.0000

Data Index: 257781 (C4 (Web Text))

Max Activating Token Index: 392

Text #12

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.6569. Min Act: -0.0000

Data Index: 268438 (C4 (Web Text))

Max Activating Token Index: 756

Text #13

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.6499. Min Act: -0.0000

Data Index: 175197 (C4 (Web Text))

Max Activating Token Index: 62

Text #14

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.8897. Min Act: -0.0000

Data Index: 973230 (C4 (Web Text))

Max Activating Token Index: 259

Text #15

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.6202. Min Act: -0.0000

Data Index: 313006 (C4 (Web Text))

Max Activating Token Index: 40

Text #16

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.4916. Min Act: -0.0000

Data Index: 900697 (C4 (Web Text))

Max Activating Token Index: 140

Text #17

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.5954. Min Act: -0.0001

Data Index: 47375 (C4 (Web Text))

Max Activating Token Index: 382

Text #18

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.4266. Min Act: -0.0000

Data Index: 558595 (C4 (Web Text))

Max Activating Token Index: 271

Text #19

Max Range: 3.6587. Min Range: -3.6587

Max Act: 2.9417. Min Act: -0.0000

Data Index: 10645 (C4 (Web Text))

Max Activating Token Index: 681