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

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

Neuron 2802 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.6499. Min Act: -0.0000

Data Index: 1452148 (Python Code)

Max Activating Token Index: 170

Text #1

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4429. Min Act: -0.0000

Data Index: 190480 (C4 (Web Text))

Max Activating Token Index: 687

Text #2

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4393. Min Act: -0.0000

Data Index: 308329 (C4 (Web Text))

Max Activating Token Index: 641

Text #3

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4577. Min Act: -0.0000

Data Index: 1252967 (C4 (Web Text))

Max Activating Token Index: 586

Text #4

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4127. Min Act: -0.0000

Data Index: 1480642 (Python Code)

Max Activating Token Index: 154

Text #5

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4575. Min Act: -0.0000

Data Index: 434055 (C4 (Web Text))

Max Activating Token Index: 833

Text #6

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3371. Min Act: -0.0000

Data Index: 1427325 (Python Code)

Max Activating Token Index: 292

Text #7

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4584. Min Act: -0.0000

Data Index: 1362780 (Python Code)

Max Activating Token Index: 390

Text #8

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4828. Min Act: -0.0000

Data Index: 307858 (C4 (Web Text))

Max Activating Token Index: 178

Text #9

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3970. Min Act: -0.0000

Data Index: 1382910 (Python Code)

Max Activating Token Index: 943

Text #10

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3897. Min Act: -0.0000

Data Index: 301799 (C4 (Web Text))

Max Activating Token Index: 392

Text #11

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4137. Min Act: -0.0000

Data Index: 1616361 (Python Code)

Max Activating Token Index: 671

Text #12

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3706. Min Act: -0.0000

Data Index: 1179825 (C4 (Web Text))

Max Activating Token Index: 843

Text #13

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4080. Min Act: -0.0000

Data Index: 1373406 (Python Code)

Max Activating Token Index: 354

Text #14

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3955. Min Act: -0.0000

Data Index: 887727 (C4 (Web Text))

Max Activating Token Index: 722

Text #15

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3590. Min Act: -0.0000

Data Index: 1171693 (C4 (Web Text))

Max Activating Token Index: 789

Text #16

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3989. Min Act: -0.0000

Data Index: 750360 (C4 (Web Text))

Max Activating Token Index: 642

Text #17

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4230. Min Act: -0.0000

Data Index: 554961 (C4 (Web Text))

Max Activating Token Index: 465

Text #18

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.3074. Min Act: -0.0000

Data Index: 686300 (C4 (Web Text))

Max Activating Token Index: 806

Text #19

Max Range: 0.6499. Min Range: -0.6499

Max Act: 0.4020. Min Act: -0.0000

Data Index: 199158 (C4 (Web Text))

Max Activating Token Index: 85