Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1594 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.0617. Min Range: -1.0617

Max Act: 1.0617. Min Act: -0.0001

Data Index: 397572 (C4 (Web Text))

Max Activating Token Index: 693

Text #1

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8656. Min Act: -0.0001

Data Index: 404919 (C4 (Web Text))

Max Activating Token Index: 82

Text #2

Max Range: 1.0617. Min Range: -1.0617

Max Act: 1.0373. Min Act: -0.0001

Data Index: 716932 (C4 (Web Text))

Max Activating Token Index: 570

Text #3

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.9717. Min Act: -0.0001

Data Index: 591242 (C4 (Web Text))

Max Activating Token Index: 995

Text #4

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8812. Min Act: -0.0001

Data Index: 664002 (C4 (Web Text))

Max Activating Token Index: 720

Text #5

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8414. Min Act: -0.0001

Data Index: 1201452 (C4 (Web Text))

Max Activating Token Index: 527

Text #6

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8642. Min Act: -0.0001

Data Index: 926882 (C4 (Web Text))

Max Activating Token Index: 273

Text #7

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8918. Min Act: -0.0001

Data Index: 1040989 (C4 (Web Text))

Max Activating Token Index: 573

Text #8

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8529. Min Act: -0.0001

Data Index: 649213 (C4 (Web Text))

Max Activating Token Index: 449

Text #9

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.7940. Min Act: -0.0001

Data Index: 1605299 (Python Code)

Max Activating Token Index: 35

Text #10

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.6936. Min Act: -0.0001

Data Index: 482221 (C4 (Web Text))

Max Activating Token Index: 789

Text #11

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8421. Min Act: -0.0001

Data Index: 993027 (C4 (Web Text))

Max Activating Token Index: 901

Text #12

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.7733. Min Act: -0.0001

Data Index: 1325086 (C4 (Web Text))

Max Activating Token Index: 242

Text #13

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.7928. Min Act: -0.0001

Data Index: 1451347 (Python Code)

Max Activating Token Index: 593

Text #14

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.7405. Min Act: -0.0001

Data Index: 946705 (C4 (Web Text))

Max Activating Token Index: 992

Text #15

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8596. Min Act: -0.0001

Data Index: 393822 (C4 (Web Text))

Max Activating Token Index: 207

Text #16

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8024. Min Act: -0.0001

Data Index: 1238704 (C4 (Web Text))

Max Activating Token Index: 462

Text #17

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.7783. Min Act: -0.0001

Data Index: 257597 (C4 (Web Text))

Max Activating Token Index: 356

Text #18

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.8522. Min Act: -0.0001

Data Index: 784449 (C4 (Web Text))

Max Activating Token Index: 312

Text #19

Max Range: 1.0617. Min Range: -1.0617

Max Act: 0.7160. Min Act: -0.0001

Data Index: 359420 (C4 (Web Text))

Max Activating Token Index: 583