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

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

Neuron 506 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: 0.0987. Min Range: -0.0987

Max Act: 0.0987. Min Act: -0.0001

Data Index: 92711 (C4 (Web Text))

Max Activating Token Index: 356

Text #1

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0985. Min Act: -0.0001

Data Index: 1636406 (Python Code)

Max Activating Token Index: 760

Text #2

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0839. Min Act: -0.0001

Data Index: 80978 (C4 (Web Text))

Max Activating Token Index: 1002

Text #3

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0811. Min Act: -0.0001

Data Index: 1629057 (Python Code)

Max Activating Token Index: 262

Text #4

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0781. Min Act: -0.0001

Data Index: 47744 (C4 (Web Text))

Max Activating Token Index: 943

Text #5

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0780. Min Act: -0.0001

Data Index: 144179 (C4 (Web Text))

Max Activating Token Index: 779

Text #6

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0797. Min Act: -0.0001

Data Index: 515501 (C4 (Web Text))

Max Activating Token Index: 857

Text #7

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0772. Min Act: -0.0001

Data Index: 419514 (C4 (Web Text))

Max Activating Token Index: 793

Text #8

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0754. Min Act: -0.0001

Data Index: 1446216 (Python Code)

Max Activating Token Index: 908

Text #9

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0698. Min Act: -0.0001

Data Index: 1206119 (C4 (Web Text))

Max Activating Token Index: 543

Text #10

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0715. Min Act: -0.0001

Data Index: 599060 (C4 (Web Text))

Max Activating Token Index: 628

Text #11

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0765. Min Act: -0.0001

Data Index: 1454388 (Python Code)

Max Activating Token Index: 887

Text #12

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0726. Min Act: -0.0001

Data Index: 1539391 (Python Code)

Max Activating Token Index: 586

Text #13

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0720. Min Act: -0.0001

Data Index: 1645119 (Python Code)

Max Activating Token Index: 46

Text #14

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0680. Min Act: -0.0001

Data Index: 1562727 (Python Code)

Max Activating Token Index: 826

Text #15

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0664. Min Act: -0.0001

Data Index: 704249 (C4 (Web Text))

Max Activating Token Index: 459

Text #16

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0703. Min Act: -0.0001

Data Index: 650153 (C4 (Web Text))

Max Activating Token Index: 800

Text #17

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0700. Min Act: -0.0001

Data Index: 476634 (C4 (Web Text))

Max Activating Token Index: 1014

Text #18

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0707. Min Act: -0.0001

Data Index: 1484919 (Python Code)

Max Activating Token Index: 830

Text #19

Max Range: 0.0987. Min Range: -0.0987

Max Act: 0.0677. Min Act: -0.0001

Data Index: 564961 (C4 (Web Text))

Max Activating Token Index: 923