Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 784 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0657. Min Act: -0.0005

Data Index: 789477 (C4 (Web Text))

Max Activating Token Index: 255

Text #1

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0638. Min Act: -0.0005

Data Index: 178276 (C4 (Web Text))

Max Activating Token Index: 622

Text #2

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0606. Min Act: -0.0005

Data Index: 303465 (C4 (Web Text))

Max Activating Token Index: 164

Text #3

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0599. Min Act: -0.0005

Data Index: 333368 (C4 (Web Text))

Max Activating Token Index: 238

Text #4

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0589. Min Act: -0.0005

Data Index: 397251 (C4 (Web Text))

Max Activating Token Index: 821

Text #5

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0588. Min Act: -0.0005

Data Index: 730550 (C4 (Web Text))

Max Activating Token Index: 575

Text #6

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0576. Min Act: -0.0005

Data Index: 1088812 (C4 (Web Text))

Max Activating Token Index: 666

Text #7

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0593. Min Act: -0.0005

Data Index: 1646446 (Python Code)

Max Activating Token Index: 507

Text #8

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0611. Min Act: -0.0005

Data Index: 211200 (C4 (Web Text))

Max Activating Token Index: 991

Text #9

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0561. Min Act: -0.0005

Data Index: 205264 (C4 (Web Text))

Max Activating Token Index: 577

Text #10

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0528. Min Act: -0.0005

Data Index: 774074 (C4 (Web Text))

Max Activating Token Index: 692

Text #11

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0563. Min Act: -0.0005

Data Index: 708468 (C4 (Web Text))

Max Activating Token Index: 1007

Text #12

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0551. Min Act: -0.0005

Data Index: 622018 (C4 (Web Text))

Max Activating Token Index: 950

Text #13

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0588. Min Act: -0.0005

Data Index: 128405 (C4 (Web Text))

Max Activating Token Index: 153

Text #14

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0551. Min Act: -0.0005

Data Index: 624831 (C4 (Web Text))

Max Activating Token Index: 913

Text #15

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0554. Min Act: -0.0005

Data Index: 600294 (C4 (Web Text))

Max Activating Token Index: 761

Text #16

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0548. Min Act: -0.0005

Data Index: 264596 (C4 (Web Text))

Max Activating Token Index: 407

Text #17

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0573. Min Act: -0.0005

Data Index: 1296265 (C4 (Web Text))

Max Activating Token Index: 945

Text #18

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0561. Min Act: -0.0005

Data Index: 790151 (C4 (Web Text))

Max Activating Token Index: 114

Text #19

Max Range: 0.0657. Min Range: -0.0657

Max Act: 0.0530. Min Act: -0.0005

Data Index: 1087449 (C4 (Web Text))

Max Activating Token Index: 103