Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 58 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0952. Min Act: -0.0003

Data Index: 918865 (C4 (Web Text))

Max Activating Token Index: 663

Text #1

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0924. Min Act: -0.0002

Data Index: 535983 (C4 (Web Text))

Max Activating Token Index: 972

Text #2

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0816. Min Act: -0.0003

Data Index: 802852 (C4 (Web Text))

Max Activating Token Index: 240

Text #3

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0831. Min Act: -0.0002

Data Index: 783048 (C4 (Web Text))

Max Activating Token Index: 603

Text #4

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0759. Min Act: -0.0004

Data Index: 76497 (C4 (Web Text))

Max Activating Token Index: 417

Text #5

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0778. Min Act: -0.0002

Data Index: 879448 (C4 (Web Text))

Max Activating Token Index: 516

Text #6

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0757. Min Act: -0.0002

Data Index: 288480 (C4 (Web Text))

Max Activating Token Index: 38

Text #7

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0786. Min Act: -0.0003

Data Index: 1167157 (C4 (Web Text))

Max Activating Token Index: 822

Text #8

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0776. Min Act: -0.0003

Data Index: 609877 (C4 (Web Text))

Max Activating Token Index: 707

Text #9

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0736. Min Act: -0.0003

Data Index: 607999 (C4 (Web Text))

Max Activating Token Index: 536

Text #10

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0801. Min Act: -0.0003

Data Index: 94087 (C4 (Web Text))

Max Activating Token Index: 325

Text #11

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0757. Min Act: -0.0003

Data Index: 866671 (C4 (Web Text))

Max Activating Token Index: 734

Text #12

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0721. Min Act: -0.0002

Data Index: 399964 (C4 (Web Text))

Max Activating Token Index: 903

Text #13

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0743. Min Act: -0.0002

Data Index: 1359301 (C4 (Web Text))

Max Activating Token Index: 605

Text #14

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0727. Min Act: -0.0003

Data Index: 832635 (C4 (Web Text))

Max Activating Token Index: 620

Text #15

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0717. Min Act: -0.0002

Data Index: 615343 (C4 (Web Text))

Max Activating Token Index: 95

Text #16

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0725. Min Act: -0.0002

Data Index: 795189 (C4 (Web Text))

Max Activating Token Index: 143

Text #17

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0737. Min Act: -0.0002

Data Index: 172844 (C4 (Web Text))

Max Activating Token Index: 442

Text #18

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0712. Min Act: -0.0003

Data Index: 1448840 (Python Code)

Max Activating Token Index: 629

Text #19

Max Range: 0.0952. Min Range: -0.0952

Max Act: 0.0726. Min Act: -0.0002

Data Index: 121761 (C4 (Web Text))

Max Activating Token Index: 363